Background Art In modern optoelectronics a multitude of new devices and methods are known. It is believed that the existing devices and methods can be further improved.

Summary of the Invention The present invention deals with optical fiber/optical component coupling assembl, method and apparatus of color mixing in a laser diode system, a system and method of frequency multiplication in a semiconductor laser diode, optical fiber/component assembly, a laser-diode assembly with external bragg grating for narrow-bandwidth light and a method of narrowing linewidth of the spectrum, opto-electronic interface module for high speed communication systems and method of assembling thereof.

Brief Description of the Drawings Fig. 1A is a partially sectional top view of an optical fiber/optical component assembly of the invention.

Fig. 1 B is an exploded three-dimensional view of the device shown in Fig. 1 A.

Fig. 1 C is a sectional view of the device of the invention along line II-II of Fig.

1A.

Fig. 1 D is a side view of the device of the invention.

Fig. 1 E is a partial three-dimensional view of the device according to another embodiment of the invention, where lens elements have dimensions accurately fit into a rectangular centering slot of the housing.

Fig. 1F illustrates is a partial three-dimensional view of a device made in accordance with an embodiment of the invention, in which the centering groove has a V-shaped configuration.

Fig. 2 is a view of a glass ferrule used for the preparation of spacers.

Fig. 3 is a three dimensional view of a glass or quartz plate with arrays of cylindrical aspherical lenses.

Fig. 4 is a three-dimensional view of an individual lens element prepared by the method of the invention.

Fig. 5 is a top view of the apparatus of the invention, which shows layout of the components of the optical system.

Fig. 6 is a longitudinal sectional view of a semiconductor laser diode assembly for generating red light in the apparatus of the invention.

Fig. 7 is a longitudinal sectional view of a semiconductor laser diode assembly for generating green and blue lights in the apparatus of the invention.

Fig. 8 is a block diagram of an electronic circuit for setting up and controlling intensity of light in one of monochromatic optical units shown in Figs. 5,6, and 7.

Fig. 9A is a schematic view of a frequency-doubling laser made in accordance with the invention for generating a blue light of a selected wavelengh.

Fig. 9B is a sectional view along lines 1 B-1 B of Fig. 9A.

Fig. 9C is a sectional view along lines 1 C-1 C of Fig. 9A.

Fig. 9D is a schematic view of a frequency-doubling laser similar to the one shown in Fig. 1A, but with housings and heat sinks separate for a laser/fiber coupling and for a fiber/crystal coupling in a system for generating a blue light of a selected wavelength.

Fig. 10 is a partial sectional view illustrating coupling of the laser diode with an optical fiber in the system of Fig. 9A.

Fig. 11 is a more detailed partial sectional view illustrating a non-linear frequency-doubling crystal and coupling thereof with a laser light input and blue light output.

Fig. 12A is a schematic view of a frequency-doubling laser made in accordance with the invention for generating a green light of a selected wavelength.

Fig. 12B is a schematic view of a frequency-doubling laser similar to the one shown in Fig. 4A, but with housings and heat sinks separate for a laser/fiber coupling and for a fiber/crystal coupling in a system for generating a green light of a selected wavelength.

Fig. 13 is a more detailed partial sectional view illustrating a non-linear frequency-doubling crystal and coupling thereof with a laser light input and green/blue light output.

Fig. 14 is a more detailed partial sectional view illustrating a non-linear frequency-doubling crystal and coupling thereof with a laser light input and green light output.

Fig. 15 is longitudinal sectional view of a known optical fiber/optical component assembly.

Fig. 16A is a longitudinal sectional view of an optical fiber/laser diode assembly of the invention.

Fig. 16B is a longitudinal sectional view of a fiber-to-fiber coupling assembly of the invention.

Fig. 17 is a view similar to that of Fig. 16A illustrating another method of fixation of the fiber in the adjusted position.

Fig. 18A is an exploded three-dimensional view illustrating assembling of a multiport coupler of the invention for coupling a laser diode matrix to a multiple-fiber port.

Fig. 18B is an exploded three-dimensional view illustrating a multiport coupler of the invention for coupling one multiple-fiber port to another multiple-fiber port.

Fig. 19 is a three-dimensional view of one of the components of the couplers of Figs. 18A and 18B made in the form of a bundle of interconnected tubes.

Fig. 20A is a plan view of a device for switching an optical path between optical fibers.

Fig. 20B is a sectional view along lines VIB-VIB in Fig. 20A.

Fig. 21 is a longitudinal sectional view of a laser-diode assembly of the invention.

Fig. 22 is a sectional view along the line i-t) of Fig. 21.

Fig. 23 is a fragmental sectional view on a larger scale illustrating the butt connection of the fiber with the end face of the lens element.

Fig. 24 is a view of a system of the invention with the reflecting mirror on the front facet of the laser diode.

Fig. 25 is a view of a system of the invention with the reflecting mirror on the rear end face of the microlens element of the device for coupling the laser diode to the output optical fiber.

Fig. 26A is a sectional view that illustrates coupling of the optical fiber with a miniature photodetector in accordance with the principle of the present invention.

Fig. 26B, which is a sectional view along the line 26B-26B of Fig. 26A.

Fig. 27A is a simplified plan view of a unit consisting of the interface of the invention and a substrate with a hybrid circuitry of commercially produced electrical components.

Fig. 27B is a view similar to Fig. 27A for the arrangement with a trans- impedance amplifier formed on the substrate in combination with a photodetector.

Fig. 28 is a sectional view similar to the one of Fig. 26A for an array-type interface.

Fig. 29 is a simplified block diagram illustrating electrical connections between the components of the device of the invention.

Fig. 30 is a top view of interface modules of the invention for optical and electrical components arranged in a matrix form.

Fig. 31 is a three-dimensional view of a matrix-type interface module of the present invention with pin/slot connections with a hybrid circuitry.

Best Mode of Carrying out the Invention An optical fiber/optical component device of the invention is shown in Figs.

26A and 26B, where Fig. 26A is a partially sectional top view and Fig. 26B is an exploded three-dimensional view of the device which is used to more distinctly illustrate some elements of the optical components. Fig. 26C is a sectional view along line li-li of Fig. 26, and Fig. 26D is a side view of the device.

As can be seen from the aforementioned drawings, the device of the invention as a whole is supported by a rectangular housing H, which has a longitudinal V-shaped or rectangular groove 29. This groove serves for placement and centering of the components of the device. A unit, which in general is designated by reference numeral 30, consists of two lens elements 32,34 and a spacer 36 sandwiched between them. This unit consitutes an anamorphotic objective.

As shown in Fig. 1 C, which is a sectional view of the device of the invention along line ll-ll of Fig. 1A, in the illustrated embodiment the groove 29 has a rectangular cross section and a width that ensures gap-free fit of the anamorphotic objective 30 in this groove. The parts that form the objective 30 are connected into an integral unit, e. g. , by gluing with a UV-curable epoxy glue. If necessary, they can be connected by thermal fusion. The flatness and parallelism of the end faces, as well as the aforementioned dimensions of the components that form the objective ensure self-alignment and self- centering of the components during assembling.

Each lens element comprises a rectangular, e. g. , square plate made of glass, quartz, or any other suitable optical material having flat and strictly parallel front and rear sides or end faces and a cylindrical spherical lens on the mating front sides. More specifically, the lens element 32 comprises a square plate 38 with an spherical cylindrical lens 40 on the front side or end face 42 that faces the lens element 34. The end face 42 is strictly parallel to the end face 43 (Fig. 1A) on the back side of the lens element 32. Similarly, the lens element 34 comprises a square plate 44 with an spherical cylindrical lens 46 on the front side, e. g. , on the end face 48 (Fig. 2A) that faces the lens element 32. The back side, e. g. , the end face 49 of the lens element 34 is strictly parallel to the end face 48 of this element. The spherical cylindrical lenses 40 and 46 have their longitudinal axes X-X and Y-Y, respectively, turned by 90° relative to each other. In the illustrated embodiment, the lenses 40 and 46 are made integrally with the plate-like bodies of the lens elements 32 and 34, respectively, e. g. , by chemical etching. If necessary, however, they can be produced by cutting a cylindrical body in a longitudinal direction and then gluing the half-cylinders to the end faces of the plates.

The spacer 36 is a ring-like element with a central hole 50 and two strictly parallel and flat end faces 52 (Fig. 1 B) and 54 (Fig. 1 A). The reason that the end faces, i. e. , 48,49, 52,54, 42,43 should be strictly parallel to each other is that they function as reference surfaces for assembling. Their surface condition should ensure that deviation of the lenses 40,46 from parallelism does not exceed 2 mm.

Shown on the left side of the anamorphotic objective 30 in Figs. 1A, 1B, 2D is a laser diode unit 60 which is supported by a cross beam 33 that rests on the shoulders 35 and 37 of the housing H located on opposite sides of the groove 29. The laser diode 60 is supported so that the center of its emitter (not shown) is located at a point of intersection of the longitudinal vertical plane V of symmetry of the groove 29 and the horizontal plane N of the shoulders 35 and 37 that supports the cross beam 33. The portions 35a and 37a of the shoulder surfaces near the diode 60 are metallized. The surfaces of the cross beam 33 which are in contact with the shoulders are also metalized and are in electrical contact with the output lead wires 60a and 60b of the diode. An example of such a laser diode is a 916 nm single-mode edge-emitter type laser diode produced by Perkin Elmer Co. The laser diode of this type has a 1 mm x 3 mm edge emitter. The technology of the present invention also applicable to diodes of a VCEL type with emitters on a vertical cavity.

The groove 29 of the housing H which supports all the aforementioned components functions as an aligning and centering element. The spacer 36 has a round cross section. The diameter of this round cross section is equal to the width of the groove 29. The depth K of the groove 29 is equal to the external radius of the spacer 36.

A distance R (Fig. 1D) from the emitter of the laser diode 60 to the lens element 34 is within the range of 1 to 100 mm. The shortened distance between the emitter of the laser diode 60 and the lens element improves optical coupling efficiency, as compared to the TO-can mounting where this distance is relatively large. It is important to ensure divergence of the optical beam OB (Fig. 1A) corresponding to the input aperture of the anamorphotic objective 30 for full optical coupling of the optical components.

The optical lenses 40 and 46 have the same length in the direction of their respective longitudinal axes X-X and Y-Y and this length has a magnitude that ensures gap-free snug fit of the lenses 40 and 46 in the hole 50 of the spacer 36 when the unit is assembled by sandwiching the spacer 36 between the lens elements 32 and 34 and the parts are secured together, e. g. , by an optical glue 56,58, e. g. , a UV-cured NOA-61 epoxy-type adhesive.

Located on the side of the anamorphotic objective 30 opposite to the laser diode 62 is a glass ferrule 68 with a central opening 70 and end faces 72 (Fig.

1 B) and 74 (Fig. 1A). The ferrule 68 is also positioned in the groove 29 but the rear end of the ferrule 68 projects from the groove for the purpose described below. The external diameter of the ferrule 68 is equal to that of the spacer 36, and therefore the ferrule 68 is also self-centered in the groove 29. The end face 72 of the glass ferrule 68 is strictly parallel to the end face 43 of the lens element 32 with deviation from flatness of less than 1 mm. An optical fiber 76 is inserted into the central opening 70 so that its front end face 76a has a butt connection with the rear end face 43 via a thin layer 78 of a UV- curable optically matched epoxy glue (such as NOA-61 type adhesive) which is used for attaching the ferrule 68 as well as the end face 76a of the optical fiber 76 to the end face 43 of the lens element 32. The glue can be, e. g. , The glue layer has a thickness of about 4-5 mm. The butt connection of the fiber to the flat side of the lens element ensures automatic positioning of the fiber in the device and thus simplicity and repeatability of such positioning under conditions of mass production. It is understood that reference numeral 76 designates both the core and the clad of the optical fiber which are not designated separately.

It is obvious that the optical axes of the fiber 76, the laser diode 62, and the anamorphotic objective 30 are strictly linear and coincident in all these components. The surface 43 of the lens element 32, including the lens 40 itself, has an anti-reflective optical coating 80 designed so as be index- matched with the NOA-61 optical epoxy layer 78 which, as mentioned above, has a maximum thickness of about 4-5 mm. This improves optical coupling of the lens to the fiber and eliminates mechanical mismatch that may be caused by thermal deformations. It is understood that similar anti-reflective coatings can be applied onto other mating optical surface, e. g. , on the end face 49 of the lens element 34, etc.

The end of the glass ferrule 68 that projects from the groove 29 is covered with a plastic or rubber boot 82 via an additional protective buffer 84. A small gap 86 is left between the end 74 of the glass ferrule 68 and the end face of the buffer 84. The boot 82 is fit onto the end of the glass ferrule 68 and is attached thereto, e. g. , by an epoxy glue. The function of the boot 82 is to prevent clad/core of the fiber from breaking due to bending.

The optical fiber/optical component coupling assembly of the present invention operates as follows: Afterthe light-emitting component, in this case the laser diode 60, is activated, a diverged light beam emitted by the diode propagates in the direction of the optical axis Z-Z, passes through the lens 34, a space between the lenses 34 and 32, and is transformed into a converged beam. This beam is collected in the point which is located in the center of the core on the mating end face of the optical fiber 76. The light beam then propagates through the optical fiber 76 to the output end of the fiber (not shown). Since all the components of the device are strictly aligned, the light energy is transmitted with high efficiency, i. e. , with the minimal losses.

Another embodiment of the invention is shown in Fig. 1 E, which is a partial three-dimensional view of the device. In this embodiment, the components of the device that correspond to those of the first embodiment, are designated by the same reference numerals with an addition of a prime symbol. Thus the lens element 34 of the first embodiment correspond in Fig. 1E to a lens element 34'. In the embodiment of Fig. 1 E, the lens elements 34'and 32' have a width equal to the width of the centering groove W'so that the alignment of the lens elements 34'and 32'will be performed not due to insertion of the lenses into the opening of the spacer 36', but by using the sides of the lenses as reference surfaces. For this embodiment, the fast and slow axes of the diode should be preoriented prior to assembling, since it would be impossible to rotate the lenses in the groove W'.

Fig. 1 F illustrates is a partial three-dimensional view of a device made in accordance with another embodiment of the invention, in which the centering groove has a V-shaped configuration. In this embodiment, the components of the device that correspond to those of the first embodiment, are designated by the same reference numerals with an addition of two primes ("). Thus, the device has a housing H", with a V-groove W"which is used foe centering and aligning an objective 30"that consists of a first lens element 34", a spacer 36", and a second lens element 38", a ferrule 68", etc. In this embodiment, however, the lens elements 34"and 38"should have a square-shaped configuration with squares inscribed into the circular cross section of the cylindrical spacer 36. The principle of assembling and operation of the device shown in Fig. 1 F is the same as in the previous embodiments.

In the manufacture of the device of the invention, first, a housing H (Fig. 1 B) with the groove 29 is produced. The surfaces of the shoulders 35 and 37 are metalized with electroconductive coatings 35a and 37a. Metallization can be carried out, e. g. , by sputtering silver, gold, or other non-oxidized metal. A piece of a glass tube 90 shown in Fig. 3 is produced with a predetermined outside diameter"D"and with an inner diameter"d", i. e. , a diameter of a central hole 92, that corresponds to a snug fit of the cylindrical spherical lenses 40 and 46 (Figs. 1A and 1B) in the opening 92. The glass ferrule is produced either from a rod-like workpiece in which a central opening 92 is drilled through the entire length of the rod, or from a ready-made glass tube of a required outer and inner diameters. The tube 90 is then diced into individual ring-like bodies 94a, 94b,.... cut strictly perpendicular to the longitudinal axis of the tube 90. The rings 94a, 94b,.... are then polished to ensure a predetermined width"P" (Fig. 2) and strict parallelism and flatness of both end faces of the rings. As a result, a plurality of spacers 36 (Figs. 1A and 1 B) are formed.

Plates (not shown at this stage of manufacturing), from which lens elements 32 and 32 have to be made, are then prepared, machined, and finished to specified parallelism and flatness. The plates are optically coated. The lenses are then formed on the lens plates, e. g. , by photolithography and chemical etching or by molding. As a result, plates 96 of the type shown in Fig. 3, which is a three-dimensional view, are produced with arrays of lenses 98a, 98b,... 98n, 100a, 100b,.... 100n,.... individual lenses shown in Fig. 4, which is a three-dimensional view, are then prepare by cutting the plates 96 into square segments that correspond in their shape and dimension to the lens elements 32 and 34 of Figs. 1A and 1B. It is understood that both elements 32 and 34 are identical and during assembling the lenses 98a, 98b,... 98n, 100a, 100b,.... 100n are just turned by 90° to form lenses 40 and 46 of the device of Figs. 1A and 1B. In dicing, the direction of cut lines L1, L2,... is strictly parallel to the longitudinal axes of the cylindrical spherical lenses 98a, 98b,... 98n, 100a, 100b,.... 100n,...., whereas, the direction of cut lines M1, M2,... is strictly perpendicular to the direction of lines L1, L2,.....

In assembling, the anamorphotic objective 30 is placed and self-centered in the groove 29 of the housing H. One end of the fiber 76 is inserted into the opening 70 of the ferrule 68 to position in which the end face of the fiber is located close to the front end face 78 of the ferrule. A layer of a UV-curable glue, optically matched with the material of the lens plate 32, is applied onto the rear end face 43 of the lens element 32. The ferrule 68 is brought in contact with the rear end face 43 of the lens element 32 which has an antireflective coating 80. It should be noted that at this stage of the assembling the glue is not yet cured and therefore allows for small movements of the fiber and small rotations of the objective 30 in the groove 29 required for subsequent adjustment. The crossbeam 60a is then placed onto the shoulders 35 and 37 with the metalized portions of the cross beam 60a in contact with the metalized surfaces of the shoulders. The fiber 76 has a core 77 and a main clad 79. The additional clad 84 is put onto the fiber 76 from the end of the fiber opposite to the objective 30.

In order to align the light source, i. e. , the laser diode 60 with the optical axis of the objective 30 and other components in the groove 29, a high-resolution CCD camera (not shown) is installed near the end of the fiber 76 opposite to the objective 30. The laser diode 60 is ignited, and the cross beam 33 with the laser diode 60 is moved in the direction of the axes X-X and Z-Z. The position of the diode in the plane X-Z is fixed. This allows to use a two- coordinate piezo actuator (not shown) for movement of cross beam 33 with the laser diode in the direction of the axes X-X and Z-Z. This movement is carried out while observing the intensity and dimensions of the light spot produced by the laser diode.

Since orientation of the fast and slow axes of the laser diode 60 is known, and the objective 30 was placed into the groove 29 with the preoriented position of the aforementioned axes, the final adjustment of the diode 60 with respect to the anamorphotic objective 30 is normally not required. If necessary, however, such adjustment can be carried out by rotating the objective in the groove 29. The adjusted position is fixed by gluing or soldering the cross beam 33 to the housing H.

Now the fiber 76 is shifted forward until its front end face 76a comes into contact with the end face 43 of the lens element 32 with an axis force of about 10 g. In this position the glue is UV-cured, whereby a butt connection is formed between the end face 43 of the lens element 32 and the front end of the ferrule 68 as well as with the front end 76a of the fiber. In this connection, a microscopic layer of the aforementioned cured glue having a thickness of about 2-3 mm is formed between the end 76a of the fiber and the end face 43.

In the production, deviation of the aforementioned gapfrom productto product does not exceed 1 mm.

The following are examples of dimensions of some individual elements in the fiber-lens assembly produced in accordance with the invention (it is understood that these dimensions are given only as examples and do not limit the scope of the invention): Diameter of the glass ferrule 68--from 0.8 mm to 1 mm Diameter of the central opening 70 of the ferrule 68---about 125 mm 0. 5 mm The fiber 76 used in these models was a single-mode 4 mm core fiber of PM- type produced by Fujikora Co. , Ltd.

It is understood that the fiber 76 may also may be a 9 mm core fiber. The coupling can be a single-mode or multi-mode fiber coupling.

Thus it has been shown that the invention provides an optical fiber/optical component assembly which is simple in construction, easy to manufacture, ensures automatic adjustment and good repeatability during assembling, preserves initial coupling properties under variable environmental conditions, improves heat-removing efficiency, is stable against thermal deformations, and provides minimal optical losses. The invention also provides a method of manufacturing and assembling the aforementioned assembly in a simple and inexpensive manner.

The invention has been shown and described with reference to specific embodiments, which should be construed only as examples and do not limit the scope of practical applications of the invention. Therefore any changes and modifications in materials, shapes, and designs of the components are possible, provided these changes and modifications do not depart from the scope of the patent claims. For example, the lens elements may have a round, hexagonal or any other suitable configuration. The optical system component that is connected to an optical fiber by the coupling device of the invention may not necessarily be a light source at all and may comprise, e. g., the end of another light-irradiating optical fiber. The optical components can be attached to each other by thermal fusion instead of gluing. The laser diode can be attached to he housing of the device by a holder different from the cross beam shown in the drawings. For example, the diode holder may have a tubularform. The tight-emitting component may comprise a laser diode with an amplifier that is placed between the laser diode and the first lens element.

The method of alignment described above is equally applicable for centering in a rectangular as well as in a V-shaped groove.

An optical color-mixing system of the present invention is shown schematically in Fig. 5, which shows three optical fiber assemblies that generate component lights of three basic RGB colors and a system of beam superposition on a common output coupler.

As shown in Fig. 5, the system has three semiconductor laser diode assemblies 10,12, and 14, which correspond to R, G, and B of the RGB system, respectively. Each assembly contains a semiconductor laser (10a, 12a, and 14a) that generates a light signal of an appropriate wavelength corresponding to R, G, and B, appropriate power supply unit (10b, 12b, and 14b), respectively, as well as temperature control means, which is described below.

An example of a semiconductor laser diode assembly 10 suitable for generating red light in the system of the invention is shown in Fig. 6. Such laser diode assembly is described in US Patent Application No......... filed by the same applicant on......

As shown in Fig. 6, a laser-diode assembly 10 for generating a frequency- stabilized narrow-bandwidth light comprises a light source in the form of a semiconductor laser diode 10a coupled via a first optical coupling device 18 to one end of a first optical fiber 20. The other end of this fiber is coupled to a second or an output fiber 22 via a second optical coupling device 24. The assembly is characterized by the fact that a long inner laser cavity L1 is formed by a section of the optical system between two oppositely directed mirror coatings M1 and M2. The first mirror coating M1 is applied onto the back side of the semiconductor laser diode 10a, and the second mirror coating M2 is applied, e. g. , onto a flat front side of an optical lens element 26 or onto the back side of another optical lens element 28 included into the laser-diode assembly 10. These optical lens elements 26 and 28 are parts of an optical coupling between the first and the second fibers 20 and 22, respectively. The first mirror coating M1 completely reflects the entire light incident onto this mirror coating, whereas the second mirror coating M2 reflects a major part of the light, e. g. , about 90% and passes only a small part, e. g. , 10% of the light incident onto this mirror. The Bragg grating 30, written in the fiber 20, is designed so that, in combination with the laser cavity L1, it suppresses the side modes of the wavelength bands and transforms them into the central mode of the narrow wavelength band, which can be passed through this grating. The light processed by the Bragg grating is passed through the second mirror coating M2 to the output fiber 22, while the reflected light performs multiple cycles of reflection between both mirrors M1 and M2 which thus form a laser resonator which amplifies the laser light output at the selected narrow waveband.

The laser diode assembly 10 has a housing H with a central longitudinal groove that is used for mounting and aligning optical components of the assembly, i. e. , the laser diode 10a, the couplers 18,24, etc. The housing H is used for stabilization of temperature of the optical components and is connected to a temperature control means that consists of an electrically controlled cooling device 19 (Fig. 6), such as a Peltier-type device, and an electronic temperature control unit 10c (Fig. 5). In Fig. 5, the electronic temperature control unit 10c is shown conventionally as a separate block.

Reference numeral 25 in Fig. 6 designates a heat-removing radiator.

Although the laser diode assembly 10 is used in the system of Fig. 5 specifically for generating the basic red-light component of 700 nm that corresponds to the standard of the International Commission on Illumination, it is understood that the above-described construction can be realized for a basic color of another three basic component color system.

Laser diode assemblies 12 and 14, which are intended for generating G and B components of the RGB system used in the present invention, have identical constructions and differ only by the types of laser diodes that generate light of 1092.4 nm for G, and 871.6 for B. The non-linear crystals will double the frequencies and thus produce on the output fibers wavelengths of 546.1 nm and 435.8 nm. For example, a laser diode assembly produced in a laboratory could generate 458 nm for blue by doubling 916 nm, 530 nm for green by doubling 1060 nm, and by using 635 nm as it is for red.

The construction of laser diode assemblies 12 and 14 for generation of green and blue lights is shown in Fig. 7 and is described in more detail in US Patent Application No..... filed by the same applicant on.............. Fig. 7 illustrates only one of the assemblies 12 and 14 which are structurally identical for both green and blue light generation assemblies and differ from each other only by characteristics of some components such as laser diodes, non-linear crystals, Bragg gratings, etc. Therefore, the description of the green/blue light- generation assemblies will be common for both assemblies, wherein the components of the green-light assembly will be shown without parentheses, while the components of the blue-light assembly will be shown in parentheses.

It can be seen from Fig. 7 that the assembly consists of a light source 32 (132) in the form of a semiconductor injection laser, which is coupled via an anamorphotic objective 38 (138) and via a butt-connection unit 40 (140) directly to an optical fiber 42 (142). The other end of this fiber is coupled to the output optical fiber 44 (144) via an air gap 46 (146), a first circular spherical lens element 48 (148), a first dichroic coating 50 (150) on the back side of the lens element 48 (148), a layer of a UV-curable glue 52 (152), a non-linear crystal 54 (154), a layer of a UV-curable glue 56 (156), a second dichroic coating 58 (158), a second microlens element 60 (160), and an air gap 62 (162). The aforementioned dichroic coatings 50 (150) and 58 (158) located on both sides of the non-linear crystal 54 (154) are opposite directed dichroic coatings of high reflectivity for lights of different wavelengths. The back facet of the semiconductor laser 32 (132) has a high-reflectivity coating 64 (164) for reflecting the fundamental wavelength. The section of the fiber 42 (142) with other appropriate components between the aforementioned high- reflectivity coating 64 (164) and the last dichroic coating 58 (158) in the direction of light propagation forms the so-called laser cavity L2 (L3), which contains a Bragg grating 66 (166). This grating selects a desired mode of the fundamental wavelength from the possible modes of the spectral range. By using different orientations of the non-linear crystal, it is possible to double, triple, or quadruple the wavelengths, and by replacing the crystals, it is possible to obtain lights of different colors. Such a construction makes it possible to form a very compact and efficient frequency-multiplied laserdiode.

In the system of the present invention shown in Fig. 7 the assemblies 12 and 14 produce on the output optical fibers 44 and 144 green (G) and blue (B) lights of RGB system, respectively. More specifically, the semiconductor laser diodes 32 and 132 of assemblies 12 and 14, respectively, are selected so that the light produced on the output of optical fiber 44 has a wavelength of 530 nm, and the light produced on the output of optical fiber 144 has a wavelength of 458 nm.

The laser diode assembly 12 (14) has a housing which consists of two parts H1a (H1b) and H2a (H2b) with central longitudinal grooves 70 (170) and 72 (172) that are used for mounting and aligning optical components of the assembly, i. e. , the laser diode 32 (132), the objectives 38 (138), etc. The housing parts H1a (H1b) and H2a (H2b) are used for stabilization of temperature of the optical components and are connected to electrically controlled cooling devices 41 a (141 a) and 41 b (141 b), such as Peltier-type devices, and an electronic temperature control units 12c (14c). In Fig. 5, the electronic temperature control units 12c and 14c are shown conventionally as separate blocks. In Fig. 7 reference numeral 74a (174a) and 74b (174b) designate heat-removing radiators.

As shown in Fig. 5, the aforementioned laser diode assemblies 10,12, and 14 are mounted on a common support or plate 200, which also supports a color combining assembly 202. This assembly includes a mounting plate 204 made of glass, quartz, ceramic, or another material with a low coefficient of thermal expansion. The mounting plate 204 has a longitudinal groove 206 of a rectangular cross section and two transverse grooves 208 and 210, which are perpendicular to the groove 206 and are parallel toueach other. The transverse grooves 208 and 210 also have a rectangular cross section. All three grooves 206,208, and 210 serve for precisely positioning respective beam collimating units 212, 214, and 216 for the formation of diffractionally- limited collimated red, green, and blue beams BR, BG, and BB. Respective units 212,214, and 216 consist of ferrules 212a, 214a, and 216a aligned and centered in the aforementioned grooves 206,208, and 210, ends of the output fibers 22,44, and 144 (Figs. 5,6, and 7) inserted into the central openings 212b, 214b, and 216b of respective ferrules 212a, 214a, and 216a, and microlens elements 212c, 214c, and 226c with microlenses 212d, 214d, and 226d inserted into the openings 212b, 214b, and 216b from the opposite sides of the ferrules 212a, 214a, and 216a. The end faces of output fibers 22,44, and 144 have to be located at strict distances from the respective microlenses 212d, 214d, and 226d so as to form the aforementioned diffractionally-limited collimated laser beams BR, BG, and BB.

Self-alignment of respective beam collimating units 212,214, and 216 is achieved due to precise cross-sectional dimensions of the grooves, accurate dimensions of the ferrules, and accurate mutual (parallel or perpendicular) positions of the grooves. More specifically, the diameters of the ferrules 212a, 214a, and 216a ensure precise sliding fit of the ferrules into the grooves, and the heights of the grooves are equal to the outer radii of the ferrules.

In Fig. 5, symbol Z designates an optical axis of the beam collimating unit 212 which coincides with the longitudinal axis of the groove 206, i. e. , with the direction of propagation of light from the output optical fiber 22 of the laser diode 10, and symbols X1 and X2 designate optical axes of the beam collimating units 214 and 216 which coincide with the longitudinal axes of the grooves 208 and 210, i. e. , with the directions of propagation of light from the output optical fibers of the laser diodes 12 and 14, respectively. It is understood that the optical axes X, Y, and Z lie in the plane of the upper surface of the plate 202.

Grooves 228 and 230, which are arranged at an angle of 45° to the axis Z, are cut in the plate 202 at points of intersection of the axes Xi and X2 with the axis Z. Inserted into these grooves 228 and 230 are semitransparent mirror plates 232 and 234, respectively. The mirror plate 232 has a coating 232a on the side that faces the beam-collimating unit 212 and a coating 232b on the opposite side of the plate. At the same time, these coatings reflect a small amount, e. g., 3% of the light power of the beam BR, towards a feedback circuit 256a, which is connected via a photodiode 258a and an analog/digital converter260a to a microprocessor 248, described later in connection with the description of the electronic control system. As will be described later, this feedback is used for controlling the luminescent characteristics, i. e. , the light power of the beam emitted from the laser diode assembly 10. At the same time, the coatings 232a and 232b ensure maximum possible reflection of the green light of 546.1 nm emitted from the laser diode assembly 12 with minimal losses.

An additional partially-transparent mirror 233 is installed in the optical path of the green beam BG from the output optical fiber 44 toward the mirror 232.

The mirror 233 passes the beam BG, but reflects a small amount, e. g. , 3% of the light power, towards a feedback circuit 256b, which is connected via a photodiode 258b and an analog/digital (A/D) converter 260b to the aforementioned microprocessor 248. As will be described later, this feedback is used for controlling the luminescent characteristics, i. e. , the light power of the beam emitted from the laser diode assembly 12.

The mirror plate 234 has a coating 234a on the side that faces the beam- collimating unit 212 and a coating 234b on the opposite side of the plate 234.

In combination with coatings 234a and 234b, the coatings 232a and 232b ensure passage of red light of 700 nm from the laser beam assembly 10 in the direction of the axis Z with minimal losses.

An additional partially transparent mirror 235 is installed in the optical path of the blue beam BB from the output optical fiber 144 toward the mirror 234. The mirror 235 passes the beam BB, but reflects a small amount, e. g., 3% of the light power, towards a feedback circuit 256c, which is connected via a photodiode 258c and an analog/digital (A/D) converter 260c to the aforementioned microprocessor248. Aswill bedescribedlater, thisfeedback is used for controlling the luminescent characteristics, i. e. , the light power of the beam emitted from the laser diode assembly 14.

In combination, the coatings 234a and 234b ensure passage of the red light beam that has passed through the semitransparent mirror plate 232 and the green light beam reflected from the coating 232b of the plate 232. At the same time, the coating 234b ensures maximum possible reflection of the blue light of 435.8 nm emitted from the laser diode assembly 14 with minimal losses.

Thus, downstream of the semitransparent mirror plate 234, a three- component beam BR-BG-BB consisting of beams BR, BG, and BB propagates in the direction of axis Z. Installed on the way of the three- component beam BR-BG-BB is a common output coupler 236 which consists of a self-aligned ferrule 238 positioned in the groove 206 and having a through opening 238a, a microlens element 240 with a microlens 242 inserted into one end of the opening 238a, and a common output fiber 244 of the entire system which is inserted into the opening 238a from the other side of the ferrule 238.

In order to provide the maximum intensity of the green and blue beams BG and BB for matching with the intensity of the red beam BR, the distances from respective beam collimating units 212,214, and 216 to the tlat back side 240a of the microlens element 240 should be substantially equal. The fine adjustment of the light beam intensities is performed only once during assembling of the system by micropositioning the respective beam collimating units with reference to the maximum output light signal separately for each wavelength and then fixing the units in the adjusted positions, e. g. , by glue.

As has been mentioned above, for controlling luminescent properties, i. e. , the light power, in each laser diode assembly 10,12, and 14, a portion of light energy, e. g. , about 3%, is extracted from the main beam and is sent via respective feedback circuits 256a, 256b, and 256c (Fig. 1) to a microprocessor 248.

It should be noted that the units for coupling the beams reflected from the respective mirrors 232,233, and 235 to the feedback fibers 257a, 257b, and 257c between the respective mirrors and the photodiodes 258a, 258b, and 258c are performed through the use of couplings 259a, 259b, and 259c.

These couplings are identical to beam collimating units 212,214, and 216 shown in Fig. 5. Furthermore, matching of the photodiodes 258a, 258b, and 258c, which received the feedback signals from the reflecting mirrors 232, 233, and 235, with the feedback fibers 257a, 257b, and 257c is carried out with the use of the same ferrules and lenses as those shown in connection with the beam collimating units 212,214, and 216. Thus, the same standard fiber/fiber or fiber/optical component unit is used for assembling the optical system. This simplifies assembling and reduces the cost of the system.

Fig. 8 is a block diagram of an electronic circuit for setting up and controlling intensity and chromaticity of light in one of monochromatic optical units shown in Figs. 5,6, and 7. It is understood that three such units are incorporated into the general system of Fig. 5. Since, in principle, the set up and control of the light intensity and chromaticity in any of the aforementioned units is identical, the diagram of Fig. 8 will be considered as a generalized circuit for all three component colors.

In the diagram of Fig. 8, reference numerals 10,12, 14 designate laser diode assemblies shown in Figs. 5,6, and 7. The input current which controls operation of the diodes 1 Oa (12a, 14a) of the assemblies 10,12, 14 is supplied to each of them from their respective power supply unit 10b or 12b or 14b (Figs. 5 and 9), which, in turn, is controlled from a microprocessor 248 via a digital/analog (D/A) converter 250. The microprocessor 248 is common for all three monochromatic systems. The subsequent description will be in singular, as it is common for each laser diode assembly. Examples of the microprocessor 248 suitable for the purposes of the present invention are Tl- 430-32, TI-430-33, and TI-430-34 units produced by Texas Instrument Co. , USA.

The microprocessor 248 is connected to a programming device, such as a computer 249 (also common for all three assemblies), and is connected in serial to respective microprocessors with an auxiliary 8-bit output which interfaces with a high speed 8-bit pulse-width modulating logic circuits (PWMLC) 254a, 54b, 254c which determine chromaticity of the colors based on duration of light pulses generated by laser diodes 1 Oa, 12a, 14a.. Forthis purpose, the PWMLC 254a, 254b, 254c are connected to the laser diodes for controlling the pulse duration byturning transistor circuits T1 a, T1 b, T1 c of the laser diodes ON on a rising edge of the pulse and OFF on the falling edge of the pulse. The function of these transistor circuits can be fulfilled by commercially produced laser diode drivers 10b, 12b ; 14b suitable for use in conjunction with laser diodes for causing them to operate in a pulse mode.

Thus, the microprocessors 252a, 252b, 252c in combination with the 8-bit PWMLC 254a, 254b, 254c makes it possible to obtain 2 8 (i. e. , 256) combinations of light intensity gradations in light emitted from each laser assembly via their respective output optical fibers 22,44, and 144 (Fig. 5).

Levels of currents in the laser diodes 10a, 12a, and 14a are controlled by means of appropriate power supply units, such as current sources 10b, 12b, and 14b (Fig. 8) for controlling the output optical signals of beams BR, BG, and BB (Fig. 5) so that they correspond, e. g. , to the aforementioned 72.1 : 1.4 : 1.0 ratio. The outputs of these current sources are connected to respective laser diodes 10a, 12a, and 14a, while their inputs are connected to the microprocessor 248 via respective digital/analog (D/A) converters 250a, 250b, 250c. In each respective laser-diode assembly, the current source (10b, 12b, and 14b) determines the luminescent characteristics or light power.

As has been shown above with reference to Fig. 5, each laser diode assembly 10,12, and 14 is connected to the microprocessor 248 via the feedback circuit 256a, 256b, 256c which contain a photodiodes 258a, 258b, and 258c and the 12-bit A/D converters 260a, 260b, 260c connected in series in the optical fibers 257a, 257b, 257c.

The microprocessor 248 is also connected to the aforementioned electronic temperature control unit 10c or 12c or 14c (Fig. 5) of its respective laser assembly. It should be noted that separate electronic temperature control units should be used for each housing part H1a (H2a) and H1b (H2b) in the laser assembles of the blue and green lights shown in Fig. 3. A single electronic temperature control unit will be used for the red laser diode assembly.

Reference numeral 271 (Fig. 8) designates a decoder unit which is common for all the laser diode assemblies and which makes it possible to adjust luminescent properties and chromaticity from voice commands, digital commands, etc. , e. g. , via a cellular phone or the like.

The electronic control system of the apparatus of the invention operates as follows : After the manufacturing of the apparatus of the invention shown in Figs. 5-8is completed, it is subjected to initial set-up at the manufacturer's facility. When the apparatus is turned on for this set up, the microprocessor 248 begins to operate in a self-diagnostic mode in which it checks all the parameters of the system, performs functional tests, and compares the measured parameters with those of the last set-up data. The aforementioned set up parameters include a temperature set points for the laser diode (1 Oa, 12a, and 14a) and the non-linear crystal (54,154). This self-diagnostic procedure is fulfilled under control of the computer 249.

Next step is to set up a standard color in each laser diode assembly so that a ratio of energetic brightnesses of the component beams corresponds to 72. 1 : 1.4 : 1.0. This is achieved by selecting levels of currents in appropriate power supply, units 1 Ob, 12b, and 14b (Fig. 8) for controlling the output optical signals of beams BR, BG, and BB (Fig. 5) so that they correspond to the aforementioned 72.1 : 1.4 : 1.0 ratio. At this stage the initial set up of the system is completed. It is understood that the 72.1 : 1.4 : 1.0 ratio is given only as an example which is in compliance with recommendations of the International Commission on Illumination and that any other ratio can be chosen. For example, in a laboratory model the laser assemblies generated beams of 635 nm, 530 nm, and 458 nm, which are different from those mentioned above.

Once the system is calibrated to the selected color ratio standard, the optical power level is permanently set in the microprocessor 248 of each color system. With the aging of the system, the aforementioned initial set-up is periodically adjusted by the microprocessor 248 in order to provide colors per selected standards.

The performance of the optical energy is periodically compared to the pre-set values in the microprocessor 248, and the power levels in 1 Oc, 12c, and 14c is automatically adjusted.

The operation of all three-laser diode assemblies 10,12, and 14 (Fig. 5) will mix the beams BR, BG, and BB of the component standard basic colors and will produce a white beam WB (Fig. 5) in the general output optical fiber 244.

It is understood that for obtaining any selected color shades, it is necessary to adjust levels of output optical signals in the output optical fibers 22, 44, and 144 (Fig. 1) in predetermined proportions. The levels of the output optical signals in the output optical fibers 22,44, and 144 are proportional to the level of current amplitude as controlled through optical feedbacks 258a, 258b, 258c, the A/D converters 260a, 260b, 260c, the microprocessor 248, the D/A converters 250a, 250b, 250c, and the current sources (10c, 12c, 14c), The level of the mean current can be regulated via the PWMLC 254a, 354b, 254c.

For this purpose, a command is sent from the computer 249 to the microprocessor 248, and the latter, in turn, sends a command via the auxiliary microprocessors 252a, 252b, 252c with an 8-bit output to the PWMLC 254a, 254b, 254c. The PWMLC 254a, 254b, 254c control the width of the current pulses in the laser diodes 10a, 12a, and 14a, thus defining the laser light power at the output of the laser diodes Once the levels of currents in the individual laser diodes 1 Oa, 12a, and 14 are adjusted, maintaining of these levels is ensured through the feedback circuits 256a, 256b, 256c. For this purpose, a feedback signal with an intensity of about 1 % of the total optical power is sent from the output optical fibers 22, 44,144 via the photodiodes 258a, 258b, 258c and the12-bit analog/digital (A/D) converters 260a, 260b, 260c to the microprocessor 248, where the feedback signal is compared with the initial setting. In the case of disagreement, the microprocessor 248 will adjust the level of the current in the appropriate laser diode via the microprocessors 252a, 252b, 252c and the 8 PWMLC 254a, 254b, 254c in a manner described above.

If necessary, the chromaticity and luminescent properties of the system can be adjusted via the decoder unit 271 by means of a voice command, digital command, etc. , e. g. , via a cellular phone or the like.

It can be seen that the correction of the current level via the feedback circuit 256 occurs in the position prior to the mixing portion of the circuit in direction of propagation of the light. Therefore the current correction procedure can be performed at any time during operation of the system without interruption of the operation. If necessary, this procedure can be automated.

Since the main components of the apparatus of the invention are based on microoptical elements interconnected through optical fibers, the overall dimensions of the entire apparatus of the invention can be reduced to miniature dimensions.

For the example, the entire optical and electrical system may have overall dimensions, which allow to built this system into any portable device.

Thus, it has been shown that the invention provides an apparatus for combining monochromatic color beams in a laser diode system which has miniature dimensions and therefore is suitable for use in portable devices, possesses high light power efficiency, is free of heat-generation problems, has low electrical consumption, has simple construction and is inexpensive to manufacture, ensures modulation of wavelength signals directly on a laser- diode light source, ensures mixing of a great number of different colors and color shades with adjustment of output color tones. The invention also provides a new method of color mixing in a laser diode optical system for obtaining a great variety of colors and color shades adjustable in a stepless manner.

Although the invention has been shown and described with reference to specific embodiments, it is understood that these embodiments were given only for illustrative purposes and that any changes and modifications with regard to shapes, designs, materials, and combinations thereof are possible, provided these changes and modifications do not depart from the scope of the patent claims. For example, the set-up procedure described in the specification is not necessarily performed for obtaining of only basic colors but can also be used for changing the intensities of the basic color beams and thus for obtaining different shades of the output colors. The microprocessor can be common or all three-color systems. The microprocessor is not necessarily an 8-bit processor and can be a 64-bit or any other processor. The color ratio may be selected in accordance with any other standard. Laser diode current modulation can be carried out in accordance with any waveform such as square, sinusoidal, etc. Laser diode current modulation and pulse width modulation can be performed simultaneously for optimum color mixing and for electrical power consumption from the battery of a portable system.

Figs. 9A, 9B, 9C, 10,11--Show Embodiment of the System for Conversion of Radiation into the Blue Light A frequency-multiplying laser system of the invention for generating doubled freqeuncy blue or green lights, as well as for lights of tripled and quadrupled frequencies has identical general schematics. The case of each color, however, differs by specific components that fulfill the same function, but for different wavelengths. The description given in this section with reference to Fig. 9A will relate to the generation of blue light. Fig. 9A is a schematic longtudinal sectional view of the double-frequency semiconductor laser system generating blue light. It consists of the following basic components sequentially mounted in a rectangular self-centering grooves 11 of a housing H in the direction of propagation of the light. The first component is a semiconductor injection laser 10. Such a laser can be represented, e. g. , by a laser-emitting diode or super-luminescent diode (SLD) described by Chin- Lin Chen in"Elements of Optoelectronics and Fiber Optics", IRWIN Publishers, Chicago, etc. , 1996, pp. 181-255.

The back facet 10a of the laser 10 has a high-reflectivity mirror 12 installed behind it for reflecting the fundamental wavelength of light generated by the laser 10. The front facet 10b of the laser 10 has an anti-reflective coating 14 which enables maximizing transmission and transmit the fundamental wavelength light of the laser 10 with extremely low losses. The photons emitted from the front facet 10b are effectively coupled into a single-mode polarization-maintaining fiber 16 via an optical coupler 18.

As shown in Figs. 9B and 9C, which are sectional views along lines 1 B-1 B and 1 C-1 C of Fig. 1 A, respectively, the cylindrical components of the system, such as the optical coupler 18, a nonlinear crystal 38, and other components, which are described below, are self-aligned in a rectangular groove 11. The construction of the housing H and mounting of the optical components are described in detail in our aforementioned pending patent application.

As shown in Fig. 9A, the optical coupler 18 has an anamorphotic objective 20 formed by a first lens element 22 and a second lens element 24 with a spacer 26 sandwiched between the both lens elements. The spacer has a through opening 28. As shown in Fig. 6, which is a more detailed longitudinal sectional view of the coupler 18, the lens element 22 has a front end face 17 and rear end face 23. The lens element 24 has a front end face 25 and a rear end face 19. The flat sides 23 and 25 of the lens elements 22 and 24, which face each other, have cylindrical spherical lenses 22a and 24a of crossed longitudinal directions. These lenses are snugly fit into the opposite ends of the through opening 28. The end faces of the lens elements and the spacer are flat with a high degree of flatness and are strictly parallel to each other, so that when the objective 20 is assembled by inserting the lenses into the opening, automatic self-alignment in relation to the longitudinal optical axis of the system is ensured. An essential feature of the coupling 18 is a ferrule 30 into which an optical fiber 16 is inserted with butt connection to the flat rear end face of the lens element 24.

The end of the fiber 16 opposite to the coupler 18 is inserted into a ferrule 34 of another optical coupler 36 (Fig. 9), which connects the fiber 16 with the aforementioned non-linear crystal unit 38. This connection is shown in more detail in Fig. 11, which is a partial sectional view. As can be seen from this drawing, the ferrule 34 has a through opening 39. The end of the fiber 16 is inserted into one end of this opening, while an aspheric circular microlens 40 of a plate-like microlens element 42 is inserted with a tight fit into the opposite end of the opening 39. The microlens element 42 is glued to the mating end face 44 of the ferrule 34 with a layer 46 of a UV-curable optically matched glue. The end face 56 of the fiber 16 inserted into the ferrule 34 is positioned with respect to the microlens 40 so that the image of the light spot produced by the microlens 40 from an additional light source (not shown) installed on the back side of the microlens 40 during the assembling procedure is strictly aligned with the center of the end face of the fiber 16.

The flat rear end face 58 of the microlens element 42 is glued via a layer 60 of a UV-curable glue to the front end face 62 of the non-linear crystal 38. The non-linear crystal 38 is a device that doubles the fundamental frequency"fo" of light and coverts it into the frequency"2f o". This can be an optically- nonlinear crystal of such a material as KTP (potassium tetanal phosphate), KTA (potassium tetanal arsenate), LBO (lithium triborate, CLBO (cesium lithium borate), BBO (beta-barium borate), etc. In the illustrated embodiment with the generation of blue light, it is recommended to use a KNBO3 crystal.

The flat front end face 66 of another plate-like microlens element 68 is glued via a layer 70 of a UV-curable glue to the rear end face 72 of the non-linear crystal 38. A circular aspheric microlens 74, which is formed on the flat rear side of the microlens element 68 inserted into a through opening 76 of an optical fiber ferrule 78. The rear end face 69 of the microlens element 68 is glued to the front end face 71 of the ferrule by a layer 73 of a UV-curable glue.

An output optical fiber 80 of the entire system is inserted into the end of the opening 76, which is opposite to the fiber 16. The position of the end face of the fiber 80 and its distance from the lens are determined and adjusted by the same method as for the fiber 16. Lenses 40 and 74 are optically aligned to be coincident with the optical axis of the system for efficient coupling of the photons to respective fibers.

Important functional features of the system of the invention are two dichroic coatings applied onto the opposite end faces of the microlens elements 42 and 68 located on opposite sides of the nonlinear crystal 38. A dichroic coating is a coating that can be designed to reflect selected portions of the spectrum light and to transmit some other selected portions of the spectrum light. More specifically, a high-reflectivity dichroic coating 82 is applied onto the rear end face 58 of the microlens element 42, whereas a second high- reflectivity dichroic coating 84 is applied onto the front end face 66 of the microlens element 68. The function of the first dichroic coating 82 is to pass only the fundament wavelength, e. g. , 918 nm, of the light generated by of the laser source 10 toward the nonlinear crystal 38 and to reflect the light of the other wavelength back into the crystal, whereas the function of the second dichroic coating 84 is to reflect the light of the fundamental wavelength 918 nm back into the non-linear crystal 38 and to pass the doubled-frequency light of 459 nm to the output fiber 80. The second dichroic coating 84 with high reflectivity in the fundamental wavelength closes the laser cavity L (Fig. 9) with a high gain factor.

In other words, the aforementioned laser cavity L is formed by the following components: the portion of the device of the invention that includes the aforementioned high-reflectivity coating 10a, the semiconductor laser diode 10, the high-transmission coating 14, the space between the laser 10 and the first lens element 22, the anamorphotic objective 20, the section of the fiber 16 from the front end to the rear end face 56 with a Bragg grating 86, the space between the end face 56 of the fiiber, and the second dichroic coating 84 (with all elements therebetween) in the direction of light propagation.

Another essential feature of the system of the invention is that the aforementioned Bragg grating 86 (Fig. 9A) written in the optical fiber 16 is within the aforementioned laser cavity L.

Bragg gratings are also known as distributed Bragg reflectors, which are optical fibers or other media that have been modified by modulating the longitudinal index of refraction of the fiber core, cladding or both to form a pattern. A fiber equipped with Bragg grating functions to modify the optical passband of the fiber (transmission characteristic) in such a way as to only transmit a narrow and controlled wavelength band. The distributed Bragg reflectors typically are"lossless"devices. In principle, the Bragg gratings can be used as light reflectors or as spectrum shape or mode converters. The latter feature of the Bragg grating is used in the system of the present invention.

A typical distributed Bragg reflector comprises a length of optical fiber including a plurality of perturbations in the index of refraction substantially equally spaced along the fiber length. These perturbations selectively reflect light of wavelength I equal to twice the spacing L between successive perturbations times the effective refractive index, i. e., ! =2nu L, where I is the vacuum wavelength and neff is the effective refractive index of the fiber for the mode being propagated. The remaining wavelengths pass essentially unimpeded. In the system of my invention, such a distributed Bragg grating is used in combination with the effect of the laser cavity length as a spectrum shape and mode converter for narrowing the spectrum bandwidth of the laser radiation, as well as for stabilization of the output laser diode characteristics and for gaining the light energy which is resonated within the laser cavity. By selecting an appropriate periodic spacing L between successive perturbations in the fiber with a distributed Bragg grating reflector, it becomes possible to select a mode, which is the most efficient for the operation of the laser 10 and ideal for non-linear crystal doubler phase matching. In the system of the invention, such a mode is the one with the maximum intensity in the laser radiation spectrum. At the same time, the gain of the maximum intensity mode is accompanied by the suppression of the side or undersired modes of the spectrum.

An anti-reflective coating 92 can be applied onto the front end face 17 (Fig.

10) of the first lens element 22 and an anti-reflective coating 93 can be applied onto the front end face of the optical fiber 16. Anti-reflective coatings can be applied onto the surfaces of the microlens, etc.

For efficient stable operation the KNbO3 non-linear crystal 38 is maintained at a phase matching angle through control of crystal temperature, using, e. g. , a thermoelectric cooler.

Fig. 9D shows Frequency-Doubling Laser System for Blue Lightwith Housings and Heat Sinks Separate to Two Different Functional Units Fig. 9D is a schematic view of a frequency-doubling laser similar to the one shown in Fig. 9A, but with housings and heat sinks separate for a laser/fiber coupling and for a fiber/crystal coupling in a system for generating a blue light of a selected wavelength.

Since in general the laser system of this embodiment is similar to the one described above for generation of blue light and differs from it only by the use of two separate housings and temperature control units for different functional parts of the system, components of the embodiment of Fig. 9D identical to those of Fig. 9A will be designated by the same reference numerals but with an addition of prime. For example, in the embodiment of Fig. 9D the laser diode will be designated by reference numeral 10'. Furthermore, a description of the identical parts will be omitted.

More specifically, an optical coupling unit 18'is supported by a housing K1 with a temperature control unit 15', e. g. , such as a Peltier device. On the other hand, the coupling unit that contains two couplers formed by a ferrule 34'with a lens element 42'and by a ferrule 78'with a lens element 68'in combination with a nonlinear crystal 38'between them is supported by a housing K2 with a temperature control unit 15", e. g. , such as a Peltier device.

The sub-assemblies supported by separate housings K1 and K2 are optically connected by a section of the optical fiber 16'with a Bragg grating 86'. The laser cavity of the device of this embodiment is designated by symbol L1'.

The remaining components of the system of Fig. 9D and structural elements of this embodiment are the same as in Fig. 9A.

Such an arrangement with individual control of temperature on the sub- assembly supported by the housing K1 and on a sub-assembly supported by the housing K2 makes it possible to maintain the aforementioned units at temperatures optimal fortheirfunctions. The temperature difference between both sub-assemblies can be within the range of 10 to 40°C. This is important since the components of the system have miniature dimensions and therefore their performance is very sensitive to temperature deviations. For example, the sub-assembly supported by the housing K2 can be maintained in a temperature phase-matched condition required forthe maximum performance efficiency of the frequency doubling process.

Embodiment of the Conversion of Radiation into the Green Light The embodiment described above with reference to Figs. 9A, 10, and 11 related to generation of blue light. Fig. 12A is a more detailed partial sectional view illustrating a non-linear frequency-doubling crystal and coupling thereof with a laser light input and green light output. In general, the system of Fig.

12A is similar to the one shown in Fig. 9A and differs from it by using a different semiconductor laser 110, that generates the light, e. g. , of 1060 nm wavelength, and a different nonlinear crystal 138, e. g. , a KTP crystal, and some other minor components.

More specifically, the back facet 11 Oa of the laser 110 has a high-reflectivity mirror 112 installed behind it for reflecting the fundamental wavelength of light generated by the laser 110. The front facet 110b of the laser 110 has a high transmission coating 114 which can transmit the fundamental-wavelength light of the laser 110 with extremely low losses. The photons emitted from the front facet 110b are effectively coupled into a single-mode polarization-maintaining fiber 116 via the optical coupler 118.

The optical coupler 118 has an anamorphotic objective 120 formed by a first lens element 122 and a second lens element 124 with a spacer 126 sandwiched between the both lens elements. The spacer has a through opening 128. As shown in Fig. 13, which is a more detailed longitudinal sectional view of the coupling 118, the lens element 122 has a front end face 117 and rear end face 123. The lens element 124 has a front end face 125 and a rear end face 119. The flat sides 123 and 125 of the lens elements 122 and 124, which face each other, have cylindrical spherical lenses 122a and 124a of crossed longitudinal directions. These lenses are snugly fit into the opposite ends of the through opening 128. The end faces of the lens elements and the spacer are flat with high degree of flatness and are strictly parallel to each other, so that when the objective 120 is assembled by inserting the lenses into the opening, automatic self-alignment in relation to the longitudinal optical axis of the system is ensured. The coupling 118 has a ferrule 130 into which an optical fiber 116 is inserted with butt connection to the flat rear end face of the lens element 124.

The end of the fiber 116 opposite to the coupler 118 is inserted into a ferrule 134 of another optical coupler 136, which connects the fiber 116 with a non- linear crystal unit 138. This connection is shown in more detail in Fig. 14, which is a sectional view. As can be seen from this drawing, the ferrule 134 has a through opening 139. The end of the fiber 116 is inserted into one end of this opening, while an aspheric circular microlens 140 of a plate-like microlens element 142 is inserted with a tight fit into the opposite end of the opening 139. The microlens element 142 is glued to the mating end face 144 of the ferrule 134 with a layer 146 of a UV-curable optically matched glue.

The end face 156 of the fiber 116 inserted into the ferrule 134 is positioned with respect to the microlens 140 so that the image of the light spot produced by the microlens 140 from an additional light source (not shown) installed on the back side of the microlens 140 during the assembling procedure be strictly aligned with the center of the end face 156 of the fiber 116.

The flat rear end face 158 of the lens element 142 is glued via a layer 160 of a UV-curable glue to the front end face 162 of the non-linear crystal 138. The non-linear crystal 138 is a device that doubles the fundamental frequency"fo "of light, e. g. , 1060 nm, and coverts it into the frequency"2fo", e. g. 530 nm (green light). The flat front end face 166 of another plate-like microlens element 168 is glued via a layer 170 of a UV-curable glue to the rear end face 172 of the non-linear crystal 138. A circular aspheric microlens 174, which is formed on the flat rear side of the microlens element 168 is inserted into a through opening 176 of an optical fiber ferrule 178. The rear end face 169 of the microlens element 168 is glued to the front end face 171 of the ferrule by a layer 173 of a UV-curable glue. An output optical fiber 180 of the entire system is inserted into the end of the opening 176 which is opposite to the fiber 116. The position of the end face of the fiber 180 and its distance from the lens are determined and adjusted by the same method as for the fiber 116.

The system has two dichroic coatings applied onto the opposite flat end faces of the microlens elements 142 and 168. More specifically, a high-reflectivity dichroic coating 182 is applied onto the rear end face 158 of the microlens element 142, whereas a second high-reflectivity dichroic coating 184 is applied onto the front end face 166 of the microlens element 168. The function of the first dichroic coating 182 is to transmit with high efficiency only the fundament wavelength, e. g. , 1060 nm, of the light generated by the laser source 110 toward the nonlinear crystal 138, whereas the function of the second dichroic coating 184 is to reflect the light of the fundamental wavelength 1060 nm back into the non-linear crystal 138, and then futher to the laser 110 via the fiber 116 (Fig. 4A), and to pass the doubled frequency light of 530 nm to the output fiber 180.

The portion of the device of this embodiment that includes the aforementioned high-reflectivity coating 110a (Fig. 12A), the semiconductor laser diode 110, the high-transmission coating 114, the space between the laser 110 and the first lens element 122, the anamorphotic objective 120, the section of the fiber 116 from the front end to the rear end face 156 with a Bragg grating 186, and the space between the end face 156 of the fiiber and the second dichroic coating 184 (with all elements therebetween) in the direction of light propagation form the aforementioned laser cavity L1. Another essential feature of the system of the invention is that the aforementioned Bragg grating 186 (Fig. 12A) is written in the optical fiber 16 within the aforementioned laser cavity L1.

An anti-reflective coating 192 can be applied onto the front end face 117 (Fig.

13) of the first lens element 122, and an anti-reflective coating 193 can be applied onto the front end face 189 of the optical fiber 116. Anti-reflective coatings can be applied onto the surfaces of the microlens, etc.

Fig. 12B is a schematic view of a frequency-doubling laser similar to the one shown in Fig. 12A, but with housings and heat sinks separate for a laser/fiber coupling and for a fiber/crystal coupling in a system for generating a green light of a selected wavelength.

Since in general the laser system of this embodiment is similar to the one described above for generation of green light and differs from it only by the use of two separate housings and temperature control units for different functional parts of the system, components of the embodiment of Fig. 12B identical to those of Fig. 12A will be designated by the same reference numerals but with an addition of prime. For example, in the embodiment of Fig. 12B the laser diode will be designated by reference numeral 110'.

Furthermore, a description of the identical parts will be omitted.

More specifically, an optical coupling unit 118'is supported by a housing T1 with a temperature control unit 115', e. g. , such as a Peltier device. On the other hand, the coupling unit that contains two couplers formed by a ferrule 134'with a lens element 142'and by a ferrule 178'with a lens element 168' in combination with a nonlinear crystal 138'between them is supported by a housing T2 with a temperature control unit 115", e. g. , such as a Peltierdevice.

The sub-assemblies supported by separate housings T1 and T2 are optically connected by a section of the optical fiber 116'with a Bragg grating 186'. The laser cavity of the device of this embodiment is designated by symbol L1".

The remaining components of the system of Fig. 12B and structural elements of this embodiment are the same as in Fig. 12A.

Such an arrangement with individual control of temperature on the sub- assembly supported by the housing T1 and on a sub-assembly supported by the housing T2 makes it possible to maintain the aforementioned units at temperatures optimal fortheirfunctions. The temperature difference between both sub-assemblies can be within the range of 10 to 40°C. This is important since the components of the system have miniature dimensions and therefore their performance is very sensitive to tempersture deviations. For example, the sub-assembly supported by the housing T2 can be maintained in a temperature phase-matched condition required forthe maximum performance efficiency of the frequency doubling process.

Operation of the Invention Since the systems of all embodiments operate in the same manner, the following description will be common for all embodiments and will refer to all preceeding drawings.

After the semiconductor laser 10 (110,110') is activated, a diverged light beam, e. g. , of 918 (1060 nm) wavelength emitted by the laser 10 (110,110'), propagates in the direction of the optical axis of the system, and passes through the lens 22a (122a, 122a') which collimates this beam. The collimated beam then propagates through the space between the lenses 22a (122a, 122a') and 24a (124a, 124a') and passes through the lens 24a (124a, 124a'), which transforms it into a converged beam spot which is collected in the point which is located in the center of the core on the mating end face 89 (189,189') of the optical fiber 16 (116,116') which is in butt connection with the lens element24 (124,124'). The light beam then propagates through the optical fiber 16 (116,116') to the output end of the fiber 16 (116,116'), which is fixed in the ferrule 34 (134,134'). On its way, the light passes through the distributed Bragg grating 86 (186,186'), which selects the desired wavelength by reflecting a portion of photons generated by the laser diode back into the laser diode. Such optical feedback forces the laser diode to operate at the desired selected wavelength. As a result, other side modes normally associated with the laser diode are suppressed, while the intensity of the laser radiation in the selected wavelength is increased. The intensified light of the selected mode then enters the lens 40 (140,140') of the lens element 42 (142,142') and passes to the lens element 68 (168,168') via the nonlinear crystal 38 (138,138') which doubles the fundamental frequency of a portion of light incident into the nonlinear crystal 38 (138,138'). This means that the aforementioned wavelength of 918 nm (1060 nm) in the aforementioned portion of the light will be converted into 459 nm (530 nm), i. e., will correspond to a blue (green) color of the light spectrum.

The first dichroic coating 82 (182,182') passes and transmits with high efficiency only the fundament wavelength, e. g. , 918 nm (1060 nm), of the light generated by of the laser source 10 (110,110') toward the nonlinear crystal 38 (138,138'and reflects light of the wavelength corresponding to the double frequency into the nonlinear crystal, whereas the second dichroic coating 84 (184, 184') reflects the light of the fundamental wavelength 918 nm (1060 nm) back into the non-linear crystal 38 (138,138') and passes the doubled- frequency light of 459 nm (530 nm) wavelength to the output fiber 80 (180, 180'). In other words, the non-linear crystal 38 (138, 138') doubles the frequency and passes the selected color (which is blue or green), while reflects back the fundamental infrared. The reflective surface of the second dichroic coating 66 (166, 166') for the chosen fundamental wavelength functions as a second reflective mirror for the laser cavity L (L1, L1'), which incorporates the nonlinear crystal 38 (138,138') for frequency doubling. This crystal is maintained at a temperature that ensures phase matching conditions.

After one pass, the chosen mode is reflected back into the laser cavity L (L1, L1'), whereas other modes pass through the Bragg grating 86 (186,186').

Approximately after two passes, a large amount of the photons is gained in the laser cavity L (L1, L1'). These photons are of a single mode with the wavelength, e. g., of 918 0.4 nm (compare with 918 12.5 nm) for the blue light or of 1060 0.4 nm (compare with 106025nm). Although the wavelength examples are given with specific tolerances of 0.4 nm, by selecting the design and the length of the laser cavity, it would be possible to reach tolerance as low as 0. 01 nm. This is the most fundamental advantage of the present invention, as it makes possible to obtain a very narrow line width and hence to produce a low-noise laser, i. e. , the laser with a very high signal/noise ratio. In this case, the obtained signal has already been collimated into the output coupling.

Such an arrangement makes it possible to maintain high level of light radiation power on the fundamental frequency sufficient for effective conversion of this radiation into the high harmonic, which in the illustrated embodiment is the second harmonic.

Thus, the non-linear crystal 38 (138,138') doubles the frequency whereby the wavelength of 918 nm (1060 nm) is converted, with a certain coefficient of conversion, into 459 nm (580 nm). Furthermore, since all the components of the device are strictly aligned, the light energy is converted with high efficiency, and the converted radiation is transmitted to the output fiber 80 (180,180') and then further to the destination with the minimal losses.

Thus it has been shown that the invention provides a frequency doubling laser diode which is made as a single integrated unit, has a compact design, can be easily coupled directly to an optical fiber, is simple in construction, reliable and efficient in operation, easy and inexpensive to manufacture, has a simplified temperature control circuit, is suitable for industrial use, and does not require labor-consuming manual alignment during the use.

Although the invention has been described with reference to specific embodiments, it is understood that these embodiments were given only for illustrative purposes and that any changes and modifications with regard to shapes, designs, materials, and combinations thereof are possible, provided these changes and modifications do not depart from the scope of the patent claims. For example, by changing the orientation of the nonlinear crystal with respect to the optical axis, it is possible to triple or quadruple the frequency.

The same objective can be achieved by changing the nonlinear crystal.

Therefore the term"frequency multiplication"should not be limited to "doubling". Furthermore, it is possible to arrange two or more non-linear crystals one after another in cascade arrangement and with respective dichroic coatings. In this case, the coupling will be realized in accordance with the priciple of the present invention. Such combinations make it possible to obtain deep ultraviolet light. The nonlinear crystals may be different from those mentioned in the description. The light source may comprise a superluminescent laser diode, a laser diode with an amplifier, etc. It is understood that specific wavelengths of 1060 nm and 916 nm were used only as examples for generating a green and blue lights, respectively, and that the green light can be generated within the range of 1000 to 1100 nm, whereas the blue light can be generated within the range of 900 to 990 nm.

An optical fiber/laser diode assembly of the invention is shown in Fig. 16A, which is a longitudinal sectional view of the device in accordance with one embodiment of the invention. Optical matching between the tight-emitting optical component, i. e. , an axisymmetric laser diode and a tight-transmitting optical fiber is carried out with the use of an optical coupler. The purpose of the coupler is to match the aforementioned components for transmitting light with minimal optical losses. In this drawing, reference numeral 100 designates a light source, e. g. , a laser diode with a substantially axisymmetric radiation.

The laser diode 100 is secured in a light-source holder 102 made, e. g. , of a metal, or any other material with good heat-removal properties for removing heat from the laser diode. An optical fiber 104, which has to receive light from the laser diode 100, is inserted with a tight sliding fit into the central opening 105 of a glass, quartz, or ceramic tube or ferrule 106. The front end face 106a of the ferrule is strictly perpendicular to the longitudinal axis X-X of the ferrule 106. A plate optical lens element 108 having a circular aspheric lens 110 is attached, e. g., by a layer 112 of a UV-cured glue, to the end face 106a of the ferrule. The base diameter of the lens 110 is equal to the diameter of the central opening 105 of the ferrule, so that the lens 110 is snugly fit into the opening 105. The front side 108a (except for the area occupied by the lens 110) and the back side 108b of the lens element 108 are polished to a high degree of surface finishing and are strictly parallel to each other and to the end face of the ferrule 106a.

It is important to note that the lens 110 and the back side 108b of the lens element are coated with antireflective coating layers 108c and 108d, respectively.

The unit consisting of the ferrule 106 with the fiber 104 and the lens element 108 attached to the ferrule 106 has to be fixed with respect to the light source, i. e. , the laser diode 100. This can be done by fixing both the unit and the holder 102 separately, or, as shown in Fig. 2A, by using a spacer 114. The front end of the spacer 114 is attached, e. g., glued, to the antireflective coating layer 108d of the lens element 108, whereas the rear end of the spacer 114 is attached to the light-source holder 102. What is meant under the term"spacer"in the present application is a an element for distancing the light source from the lens so that the aperture of the lens 110 corresponds to the entire cross section of beam. It is understood that the light spot does not have a distinct border, and therefore the term"entire cross section"means about 98% of light energy emitted from the light source.

The rear end face 106b of the ferrule 106 has butt connection to a locking ferrule 116 via a layer 118 of a UV-curable glue. The glue that forms the layer 118 is applied onto the rear end of the ferrule 106 during operation of alignment of the optical fiber 104 with the light source 100. This alignment operation will be described below. The glue also penetrates into the gap between the fiber 104 and the surface of the central opening 105 and into the opening of the locking ferrule 116. The place of butt connection is covered with a protective boot 120 made of a resilient material, e. g. , rubber or plastic.

It is important to assemble the coupling of the invention shown in Fig. 16A in such a manner that the image of the light spot produced by the lens element 108 as a result of passing the light of the source 100 through the lens 110, be strictly aligned with the center of the end face 104a of the fiber 104. It is also important that the output aperture of the lens 110 corresponds to the input aperture of the fiber 104 under condition of the minimal optical losses. This condition can be fulfilled only at a predetermined position of the end face 104a of the fiber 104 with respect to the lens element 108 and with regard to specific optical characteristics of this element. For the same purpose, i. e., decrease of optical losses, the lens is made spherical with a profile satisfying the above requirement.

The predetermined position of the end face 104a of the fiber 104 with respect to the lens element 108 is achieved as follows. The unit, consisting of the light-source holder 102 with the light source 100, the spacer 114, the lens element 108, and the ferrule 106 with the optical fiber 104 is secured in a fixture (not shown) with a photodetector 122 having a coupler for a fiber (not shown). The fiber 104 itself remains free for movements in the axial direction.

The fiber is then shifted to a position in which the size of the light spot becomes minimized. This is the position in which the fiber 104 has to be fixed with respect to the lens 110. The fiber is then moved backward for a known distance, e. g. , 0.5 mm with use of a micropositioner (not shown) while being observed under a microscope. A measured amount of a UV-curable glue is then applied onto the rear end face 106b of the ferrule 106 and onto the adjacent surface of the fiber clad projecting from the ferrule 106. While the glue is still in a highly fluid state, the fiber is shifted back to the adjusted position for aforementioned predetermined distance, and the front end face 116a is brought into butt contact with the rear end face 106b of the ferrule.

The glue penetrates to some depth into the gaps between the fiber 104 and inner surface of the central opening 105 of the ferrule 106 and into the central opening 115 of the locking ferrule 116. The glue is UV-cured, whereby a cured glue layer 118 is formed. The rubber boot 120, which is fit onto the locking ferrule 116, is pulled onto the rear end of the ferrule 106 to protect the fiber fixation area from damage by bending. As a result, the fiber 104 is reliably fixed in the ferrules 106 and 116 in a position of optical match with the lens element 108.

An optical fiber-to-fiber coupling assembly of the invention is shown in Fig.

16B, which is a longitudinal sectional view of the device in accordance with another embodiment of the invention. Optical matching between the light- emitting optical components, i. e. , a light-irradiating optical fiber and a light- transmitting optical fiber, is carried out with the use of an optical coupler. The purpose of the coupler is to optically match the aforementioned parts for transmitting light with minimal optical losses. In this drawing, reference numeral 100"designates a light source, e. g. , an optical fiber the end of which irradiates light, generated, e. g. , by a semiconductor laser (not shown in this embodiment). The light-irradiating fiber 100"is supported by a glass ferrule 102". An optical fiber 104", which has to transmit the light received from the light-irradiating optical fiber 100", is inserted with a tight sliding fit into the central opening 105"of a glass, quartz, or ceramic tube or ferrule 106". The front end face 106a"of the ferrule is strictly perpendicular to the longitudinal axis X"-X"of the ferrule 106". A plate optical lens element 108"having a circular aspheric lens 110"is attached, e. g. , by a layer 112"of a UV-cured glue, to the end face 106a"of the ferrule. The base diameter of the lens 110" is equal to the diameter of the central opening 105"of the ferrule, so that the lens 110"is snugly fit into the opening 105". The front side 108a" (except for the area occupied by the lens 110") and the back side 108b"of the lens element 108"are polished to a high degree of surface finishing and are strictly parallel to each other and to the end face 106a"of the ferrule.

It is important to note that the lens 110"and the back side 108"b of the lens element are coated with antireflective coating layers 108c"and 108d", respectively.

The unit consisting of the ferrule 106"with the fiber 104"and the lens element 108"attached to the ferrule 106"has to be fixed with respect to the light - irradiating fiber 100". The front end of the ferrule 102"is attached, e. g., glued, to the antireflective coating layer 108d"of the lens element 108", whereas the rear end face 102a"of the ferrule 102"is attached to the front end face 114a of a locking ferrule 114"., e. g. , by a layer of a UV-curable glue 116". The end face 100a"of the light-irradiating optical fiber 100"is spaced from the lens 110"so that the aperture of the lens 110"corresponds to the entire cross section of beam. It is understood that the light spot does not have a distinct border and therefore the term"entire cross section"means about 98% of light energy emitted from the light source.

The rear end face 106b"of the ferrule 106"has butt connection to a locking ferrule 120"via a layer 118"of a UV-curable glue. The glue that forms the layer 118"is applied onto the rear end of the ferrule 106"during operation of alignment of the optical fiber 104"with the light-irradiating optical fiber 100".

The glue also penetrates into the gap between the fiber 104"and the surface of the central opening 105"and into the opening of the locking ferrule 120".

It is important to assemble the coupling of the invention shown in Fig. 16B in such a manner that the image of the light spot produced by the lens element 108"as a result of passing the light of the source 100"through the lens 110", be strictly aligned with the center of the end face 104a"of the fiber 104". It is also important that the output aperture of the lens 110"corresponds to the input aperture of the fiber 104"under condition of the minimal optical losses.

This condition can be fulfilled only at a predetermined position of the end face 104a"of the fiber 104"with respect to the lens element 108"and with regard to specific optical characteristics of this element. For the same purpose, i. e., decrease of optical losses, the lens is made spherical with a profile satisfying the above requirement.

The predetermined position of the end face 104a of the fiber 104 with respect to the lens element 108 is achieved as has been described above with reference to Fig. 16A.

The coupling assemblies of the embodiments shown in Figs. 16A and 16B make it possible to achieve coefficients of optical coupling which are close to maximal theoretical values of about 98%. This is unattainable with constructions that utilize GREEN lenses or ball lenses.

Fig. 17 is a view similar to that of Fig. 16A illustrating another method of fixation of the fiber in the adjusted position. In general, this embodiment is similar to the one described with reference to Fig. 16A. The parts similar to those of Fig. 16A are designated in Fig. 17 by the same reference numerals with an addition of a prime symbol. Thus the device has a light source 100' in a light-source holder 102', etc. The embodiment of Fig. 3 differs from the embodiment of Fig. 16A in that it does not have a locking ferrule such as the locking ferrule 116 of the previous embodiment, and fixation of the fiber 104' with respect to the lens element 108'is achieved by forming a through hole 130 in the wall of the ferrule 106'which is filled with a drop of a UV-curable glue 132 added after the position of the fiber has been adjusted and has to be fixed.

Fig. 18A is an exploded three-dimensional view illustrating assembling of a multiport coupler of the invention for coupling a laser diode matrix to a multiple-fiber port. The device is suitable for mass production. The coupler consists of a light source holder 200 or a light source matrix, e. g. , 32 x 32 VCSEL's emitter (VCSEL-Vertical Cavity Surface Emitting Laser), or a similar light diode matrix, which supports arrays 202 and 204 of light sources, such as laser diodes, respectively, only two of which, i. e. , 202a, 202b can be seen in Fig. 18A. It is understood that 2x2 matrix is shown only as an example of multiplicity and that greater amount of arrays with a greater amount of diodes in each array can be used.

It is assumed that a laser diode matrix is a commercially produced unit with predetermined pitch (center to center spacing) between the emitters of individual laser diodes 202a, 202b, etc. Such diode matrices or arrays are produced, e. g. , by Hitachi, Sony, Lucent, etc.

Next in the direction of propagation of light from the diodes is a spacer 206, which holds the diode matrix 200 at a distance from a microlens matrix 208 located behind the spacer 206 in the light propagation direction. It is understood that arrangement of microlenses in this matrix corresponds to the aforementioned arrangement of the diodes in the commercially produced diode matrix 200. The second important function of the spacer 206 is removal of heat from the optical elements. For this purpose, the spacer 206 is made of a dielectric material with good heat-conductive properties, such as a glassy carbon.

The spacer 206 can be made in the form of a plate with openings 206a, 206b, etc. for passing the light beams from the laser diodes 202a, 202b, etc. The spacer 206 can be made in the form of a frame.

The microlens matrix 208 is made in the form of a plate with arrays 210 and 212 of microlenses 21 Oa, 21 Ob and 212a, 212b, respectively. It is understood that a 2x2 microlens matrix is shown only as an example and that a larger amount of the arrays, as well as the lenses in each array, can be used.

The arrangement of the microlenses 210a, 210b and 212a, 212b in the microlens matrix 210 is strictly matched with the arrangement of the laser diodes 202a, 202b, etc. , respectively, in the laser diode matrix 202. In other words, in an assembled unit, each laser diode should be strictly coaxial with the respective microlens along the respective optical axes of the system. All individual axes passing through each pair of matched laser diodes and microlenses are strictly parallel to each other.

The front end face 200c of the laser diode holder 200 (what is meant in this context under the term"front end"is a vertical plane in which the surfaces of the emitters of all laser diodes are located), the mating end face 206b and the opposite end face 206c of the spacer 206, as well as both end faces 208c and 208d of the microlens matrix 208 are strictly parallel to each other and finished to a high degree of flatness.

The next component of the assembly in the light propagation direction is a plate-like fiber termination matrix or holder 214 with arrays 216 and 218 of through openings 216a, 216b and 218a, 218b, respectively. The arrangement of these openings is strictly matched with the aforementioned patterns of laser diodes 202a, 202, etc., respectively, in the laser diode matrix 202. In other words, in an assembled unit, the longitudinal axis of each opening 216a, 216b and 218a, 218b should be strictly coaxial with the respective laser diodes 202a, 202b, etc. , and with microlenses along the respective optical axes of the system.

The fiber termination holder 214 has both end faces 214a and 214b strictly parallel and finished to a high degree of flatness. The holes 216a, 216b and 218a, 218b have diameters that correspond to the base diameters of respective lenses 210a, 210b and 212a, 212b, so that when the end face 208d of the microlens matrix 208 is brought into contact with the mating end face 214a of the fiber termination holder 214, the lenses enter the respective holes with a tight fit.

The last component of the system of the invention in the direction of light propagation is the locking plate 220 which has arrays 222 and 224 of holes 222a, 222b and 224a, 224b which are coaxial to the aforementioned lenses 210a, 210b and 212a, 212b and holes 216a, 216b and 218a, 218b.

Reference numerals 219a, 219b, 221 a, and 221 b designate light-receiving fibers inserted into the holes 222a, 222b and 224a, 224b of the locking plate 220.

In order to ensure the aforementioned coaxiality of all respective lenses in the microlens matrix 208 with holes of the fiber termination holder 214 and of the locking plate 220, the matrix 208, the holder 214, and the locking plate 220 are manufactured with the use of a common set of photomasks (not shown) for a photolithography process. The photolithography can be used for manufacturing microlenses 210a, 210b and 212a, 212b, as well as for the openings 216a, 216b and 218a, 218b of the fiber termination holder 214 and for openings 222a, 222b, and 224a, 224b of the locking plate 220. In the stepwise manufacturing procedure, coaxiality of the lenses and respective holes is achieved due to the use of at least three alignment marks on each component, i. e. , marks 226a, 226b, 226c on the microlens matrix 208, marks 228a, 228b, 228c on the fiber termination holder 214, and marks 230a, 230b, 230c on the locking plate 220. The locking plate 220 has a front end face 220a and a rear end face 220b.

The aforementioned processes of photolithography and--alignment for manufacturing lens arrays and arrays of matched optical components are known in the art and described, e. g. , in US Patent No. 6,055, 107 issued in 2000 to Yeh-Tseng Li, et al., US Patent No. 5,997, 756 issued in 1999 to Yuichi Okazaki, et al., US Patent No. 5,948, 281 issued in 1999 to Yuichi Okazaki, US Patent No. 5,871, 888 issued in 1999 to Paul Heremans, et al., and US Patent No. 5,867, 321 issued to Ken'ichi Nakama et al.

The aforementioned microlenses and holes can be produced not necessarily by photography. For example, the holes can be ultrasonically drilled or drilled on a coordinate drilling machine of high accuracy, while microlenses can be molded from electrolytically prepared molds. However, even the case of mechanical manufacturing, marks of alignment are applied by photolithography, e. g. , with the use of common mask for producing marks of alignment. Alignment holes can be pin holes.

Fig. 18B is an exploded three-dimensional view illustrating a multiport coupler of the invention for coupling one multiple-fiber port to another multiple-fiber port. In general, the device of Fig. 18B is similar to the device of Fig. 18A and differs from it by the fact that optical coupling is carried out between two multiple-fiber ports, where one of the ports functions as a light irradiating port, while the second port functions as a light receiving port. More specifically, the device has a tight-emitting multiple-fiber port 200'which comprises a plate-like body 201'with through holes 202a', 202b', 204a', and 204b'for insertion and fixation of respective light input optical fibers 203a', 203b', 205a'and 205b' which can be connected to respective light sources such as laser diodes (not shown).

The end face of respective fibers are spaced at a predetermined distance from the front end face 201 a'of the plate-like body 201'. This distance functions as the spacer 206 of the device shown in Fig. 18A.

The next component in the direction of light propagation from the light-emitting multiple-fiber port 200'is a microlens matrix 208'with microlenses 210a', 21 Ob', 212a', etc. , formed on the front end face 208a'of the microlens matrix 208'.

Next optical component in the light propagation direction is a fiber termination holder 214'with through holes 216a', 216b'and 218a', 218b'for positioning and fixation of light-receiving fibers 219a', 219b', 221 a', and 221 b'.

It is understood that the respective fibers, holes, and lenses in optically coupled components of the device of Fig. 4B should coaxial and aligned. The alignment can be facilitated by using the same alignment marks that have been described in connection with the embodiment of Fig. 18A.

Fig. 19 is a three-dimensional view of a fiber termination holder 300 made in the form of a bundle of interconnected tubes 302a, 302b and 304a, 304b. The purpose and principle of use of the fiber termination holder 300 is the same as in the case of the plate-like holder 214 shown in Fig. 18A. However, the manufacturing process does not require the use of masks for photolithography, since the bundles is produced by selecting tubes of equal inner and outer diameters with subsequent tight interconnection of the tubes, e. g. , by gluing, into a bundle and machining the opposite ends of the bundles.

If the number of fibers is small, e. g. , 4 to 16, assembling of the multiport coupling unit of the type shown in Fig. 18A and 18B can be carried out in the same manner as has been described with reference to a single-fiber coupler of Fig. 16A by adjusting each fiber individually. However, under conditions of mass production or when the number of fibers exceeds 36,64, or more, the assembling and alignment are carried out as follows.

First, the front end face of the lens termination holder 214 (214') is glued or otherwise connected to the mating end face 208d (208d') of the microlens matrix 208 (208'). Since the base diameter of the microlenses corresponds to the diameter of holes in the lens termination holder 214 (214'), the lenses will fit into these holes, and due to parallelism and flatness of the mating end faces of the holder 214 (214') and the microlens matrix 208 (208'), theses components will be aligned automatically.

The diode matrix 200 (Fig. 18A) is secured in the spacer 206, and the spacer 206 with the diode matrix is then fixed in a fixture (not shown). A sub- assembly consisting of the microlens matrix 208 (208') (Figs. 18A and 18B) and the fiber termination holder 214 (214') is installed in a position where the microlenses are aligned with the individual laser diodes (tight-emitting fibers 203a', 203b', etc. ). For the case of Fig. 4A, this is achieved by bringing the spacer 206 and the diode matrix 200 in contact with each other, igniting the laser diodes, and passing the light through the holes of the aforementioned subassembly toward a tight-intensity evaluating means such as, e. g. , a screen or a high-resolution CCD camera (not shown) which is installed directly behind the unit. The spacer 206 or the diode matrix 200 (with the microlens matrix being already attached thereto) are then moved relative to each other by means of a micropositioner in two mutually perpendicular directions in a plane perpendicular to the light propagation direction. The adjustment is carried out while observing the image of light spots on the aforementioned screen or the CCD camera. The alignment is achieved when these spots have regular forms and uniform light intensity distribution over the images of the holes. In this position the spacer 206 is secured to the diode matrix, e. g. , by a UV- curable glue. Now the subassembly consists of four aligned components, i. e., the diode matrix 200, the spacer 206,, the microlens matrix 208, and the fiber termination holder 214.

In the case of the embodiment of Fig. 18B, the alignment procedure is the same, but manipulations are performed with the multiple-fiber port 200', and the tight emitted from the ends of the fibers 203a', 203b', etc. , is generated by the laser diode which are not shown in Fig. 18B. the ends of these fibers are fixed in the holes 202a', 202b', etc. , of the plate 201'.

The next assembling step is insertion, aligning, and fixation of individual fibers 219a, 219b (219a', 219b'), and 221 a, 221 b (221 a', 221 b') in positions optically matched with respective diodes (tight-emitting fibers), lenses, and holes. For this purpose, optical fibers 219a, 219b (219a', 219b'), and 221 a, 221 b (221 a', 221 b') are passed through the holes 222a, 222b and 224a, 224b (222a', 222b' and 224a', 224b') to protrude for a distance, which approximately is equal to the thickness of the fiber termination holder 214 (214'). A plate (not shown) is then installed in front of the fiber ends strictly parallel to the front end face 220a (220a') of the locking plate 220 (220'). The fibers are shifted forward to contact the aforementioned plate, whereby after removal of the plate, the ends of the fiber are located strictly in one plane and at an equal distance from the microlenses. The fiber are the fixed to the locking plate 220 (220') by glue applied to the rear end face 220b (220b') of the locking plate 220 (220'). The unit of the plate 220 (220') with fibers is moved towards the fiber termination holder 214 (214') so that the ends of the fibers 219a, 219b (219a', 219b'), and 221 a, 221 b (221 a', 221 b') are tightly fitted into the respective openings 216a, 216b (216a', 216b') and 218a, 218b (218a', 218b') of the holder 214 (214').

The alignment and optical matching of the fibers 219a, 219b (219a', 219b'), and 221 a, 221 b (221 a', 221 b') with respect to the microlenses 21 Oa, 21 Ob (21 Oa', 21 Ob') and 212a, 212b (212a', 212b') is achieved by installing a screen or high-resolution CCD's at the rear ends of the fibers. The laser diodes are then ignited, the light passes through the microlenses, is transmitted through the fibers, and is projected onto the aforementioned screen or CCD's. The adjustment is carried out by shifting the ends of the fibers 219a, 219b (219a', 219b'), and 221 a, 221 b (221 a', 221 b') with respect to the microlenses 21 Oa, 21 Ob (21 Oa', 21 Ob') and 212a, 212b (212a', 212b'). The maximum of the light signals on the receiving end is used for defining the final position of the fiber ends. After the positioning is completed, the sub-assembly consisting of the locking plate 220 (220') with the fibers is moved away from the rest of the unit for a known distance, e. g. , of 0.5 mm. The space between the components is filled with a UV-curable glue, and the aforementioned sub-assembly is returned to the adjusted position for fixing the components by curing the glue.

Fig. 20A and 20B illustrate an embodiment of the device of the invention with means for switching an optical path between an input optical fiber 400 and a four output optical fibers 402a, 402b, 402c, and 402d. It is understood that one input optical fiber and four output optical fibers are shown only as an example, since switching can be performed between groups of fibers on the input and output sides, or between one input and two output fibers. Fig. 20B is a sectional view along lines VIB-VIB in Fig. 20A.

The basic element of the device of Figs. 20A and 20B is a linear array 404 of microlenses 404a, 404b, 404c, and 404d. The array 404 is sliding fitted into a guide slot 406 formed in a housing 408 in a direction perpendicular to the optical axis X"'-X"'of the device. Continuation of one wall of the guide slot 406 in the vertical direction is formed by a spacer 410 in the form of a plate rigidly attached to the housing 408. An end face of a ferrule 412 is attached to the spacer 410 on the side opposite to the array 404. A central hole 414 of the ferrule and a hole 416 of the spacer 410 coaxial to the hole 414 and having the same diameter are used for insertion and fixation of the light input fiber 400. The end face of the fiber 400 is spaced from the flat end face 405a of the array 404 at a predetermined distance determined in accordance with the same principle as has been described above with respect to positioning the end faces of fibers 100"and 104" (Fig. 16B).

The microlenses 404a, 404b, 404c, and 404d are inserted with tight fit into through openings 416a, 418a, 420a, and 422a of respective ferrules 416, 418, 412, and 422 which are axially aligned with the optical axes of the respective microlenses. The end faces of the ferrules 416,418, 412, and 422 are attached, e. g. , with the use of a UV-curable glue, to the lens side of the microlens array 404.

Inserted into the ferrule opening 416a, 418a, 420a, and 422a from the side opposite to the microlenses are output optical fibers 402a, 402b, 402c, and 402d. The end faces of the output optical fibers 402a, 402b, 402c, and 402d are spaced from the respective microlenses 404a, 404b, 404c, and 404d in accordance with the same principle that was used for distancing the end face 104a"from the microlens 110"in the device of Fig. 17.

The array 406 is attached to a micropositioning mechanism 500 which consist of a lever 502 rotating on a pivot pin 504. The long arm 502a of the lever 502 has a longitudinal slot 506 for sliding a pin 508 attached to the array 406. The opposite short arm 502b of the lever 502 is attached to an actuating member 510 of a stepper micropositioner 512, such as piezotransducers produced by Polytec Co. , USA.

In operation, activation of the micropositioner 512 causes sequential or I programmed switching of the array 404 so that a selected output optical fiber of the group consisting of optical fibers 402a, 402b, 402c, and 402d is aligned and optically coupled with the input optical fiber 400.

Thus, it has been shown that the invention provides an optical fiber coupler, which is simple in construction, easy to align in assembling under conditions of mass production, and is suitable for use in optical fiber multiport couplers.

The invention also provides a method of manufacturing and assembling an optical coupler for coupling an optical fiber with an optical component, which is simple and is suitable for mass production.

Although the invention has been shown and described with reference to specific embodiments, it is understood that these embodiments should not be construed as limiting the application of the invention and that any changes and modifications are possible, provided they do not depart from the scope of the appended claims. For example, in the multiport embodiment, the lens elements may have a square, hexagonal or any other suitable configuration.

The optical system component that is connected to an optical fiber by the coupling device of the invention may not necessarily be a light source at all and may comprise, e. g. , the end of another light-irradiating optical fiber. The optical components can be attached to each other by thermal fusion instead of gluing. The spacer is not necessarily made as a separate part and may be an extension of the lens element that projects from the front end face of this element towards the laser diode holder. Similarly, the spacer can be an integral part of the laser diode holder. The spacer length can be adjustable.

It is understood that the linear array 404 of the embodiment of Figs. 6A and 6B can be replaced by a matrix of microlenses so that switching can be performed in the direction of two orthogonal axes. The switching can be performed by a mechanism different from the one shown in Figs. 6A and 6B, e. g. , with the use of MEMS (micro-electro-mechanical systems). The microswitch can be connected to the light input element or elements instead of light output optical fibers. The microswitch can be used for switching the light components on and off or for switching between the components.

A laser-diode assembly of the invention is shown in Fig 21, which is a longitudinal sectional view. The assembly as a whole is supported by a rectangular housing H, which has a longitudinal rectangular groove 20 shown in Fig. 22, which is a sectional view along the line tt-ti of Fig. 21. The housing H is connected to a heat sink 21 which may have an electric control (not shown). This groove serves for placement and centering of the components of the device. A unit, which in general is designated by reference numeral 30, consists of two lens elements 32,34 and a spacer 36 sandwiched between them. This unit constitutes an anamorphotic lens assembly.

As shown in Fig. 22, in the illustrated embodiment the groove 20 has a rectangular cross section and a width that ensures gap-free fit of the optical system components, including the anamorphotic lens assembly 30. In such positions the aforementioned components are aligned with the optical axis of the laser diode assembly. The parts that form the lens assembly 30 are connected into an integral unit, e. g. , by gluing with a UV-curable epoxy glue.

If necessary, they can be connected by thermal fusion. The flatness and parallelism of the end faces, as well as the aforementioned dimensions of the components that form the lens assembly ensure self-alignment and self- centering of the components during assembling.

Each lens element comprises a rectangular, e. g. , square plate made of glass, quartz, or any other suitable optical material having flat and strictly parallel front and rear sides or end faces and a cylindrical aspherical lens on the mating front sides.

Shown on the left side of the anamorphotic lens assembly 30 in Fig. 1 is a laser diode unit 10, which is mounted on a ceramic support 11 mounted on the housing H. The laser diode 10 is supported so that the center of its emitter 13 is located on the optical axis Z-Z of the lens assembly 30. An example of such a laser diode is a 916 nm single-mode edge-emitter type laser diode produced by Perkin Elmer Co. The laser diode of this type has a 1 mm x 3 mm edge emitter. Another example is a laser diode of produced by Hitachi Co. , Ltd. for radiating light with the wavelength of 635 nm. In fact, the principle of the invention is applicable to lasers of any types and designs that emit light with wavelengths in the range of 400 to 1600 nm. For example, the technology of the present invention also applicable to diodes of a VCSEL type with emitters on a vertical cavity.

The groove 20 (Fig. 22) of the housing H, which supports all the aforementioned components, functions as an aligning and centering element, as well as a temperature-stabilizing/heat sinking chassis of the laser diode chip and the optical assembly. The spacer 36 has a round cross section. The diameter of this round cross section is equal to the width of the groove 20.

The lens element 32 (Figs. 21 and 22) comprises a plate with an spherical cylindrical lens 40 on the front end face 42 that faces the lens element 34.

Similarly, the lens element 34 comprises a square plate with an spherical cylindrical lens 46 on the front end face 48 that faces the lens element 32.

The backside, e. g. , the end face 49 of the lens element 34 is strictly parallel to the end face 48 of this element. The spherical cylindrical lenses 40 and 46 have their longitudinal axes respectively, turned by 90° relative to each other. In the illustrated embodiment, the lenses 40 and 46 are made integrally with the plate-like bodies of the lens elements 32 and 34, respectively, e. g., by chemical etching. If necessary, however, they can be produced by cutting a cylindrical body in a longitudinal direction and then gluing the half-cylinders to the end faces of the plates.

The spacer 36 is a ring-like element with a central hole 50 and two strictly parallel and flat end faces 52 and 54. The reason that all aforementioned end faces should be strictly parallel to each other is that they function as reference surfaces for assembling. Their surface condition should ensure that deviation of the lenses 40,46 from parallelism does not exceed 2 mm.

A distance R (Fig. 21) from the emitter of the laser diode 10 to the lens element 34 is within the range of 1 to 100 mm. The shortened distance between the emitter of the laser diode 10 and the lens element improves optical coupling efficiency, as compared to the TO-can mounting where this distance is relatively large. It is important to ensure divergence of the optical beam OB1 corresponding to the input aperture of the anamorphotic lens assembly 30 for full optical coupling of the optical components.

The optical lenses 40 and 46 have the same length in the direction of their respective longitudinal axes and this length has a magnitude that ensures gap-free snug fit of the lenses 40 and 46 in the hole 50 of the spacer 36 when the unit is assembled by sandwiching the spacer 36 between the lens elements 32 and 34 and the parts are secured together, e. g. , by an optical glue, e. g. , a UV-cured NOA-61 epoxy-type adhesive. Alternatively, the parts can be secured together by means of laser welding, such as glass-to-glass YAG laser welding.

Located on the side of the anamorphotic lens assembly 30 opposite to the laser diode 10 is a glass ferrule 68 with a central opening 70 and end faces 72 and 74. The ferrule 68 is also positioned in the groove 20. The external diameter of the ferrule 68 is equal to that of the spacer 36, and therefore the ferrule 68 is also self-centered in the groove 20. The end face 72 of the glass ferrule 68 is strictly parallel to the end face 49 of the lens element 34 with deviation from flatness of less than 1 mm. An optical fiber 76 is inserted into the central opening 70 so that its front end face 76a has a butt connection with the rear end face 49 via a thin layer 78 of a UV-curable optically matched epoxy glue (such as NOA-61 type adhesive) which is used for attaching the ferrule 68 as well as the end face 76a of the optical fiber 76 to the end face 49 of the lens element 34. This is shown in Fig. 3, which is a fragmental sectional view on a larger scale illustrating the butt connection of the fiber with the end face of the lens element.

The glue layer has a thickness of about 4-5 mm. The butt connection of the fiber to the flat side of the lens element ensures automatic positioning of the fiber in the device and thus simplicity and repeatability of such positioning under conditions of mass production. It is understood that reference numeral 76 designates both the core and the clad of the optical fiber, which are not designated separately.

It is obvious that the optical axes of the fiber 76, the laser diode 10, and the anamorphotic lens assembly 30 are strictly linear and coincident in all these components.

The flat surface 43 (Fig. 23) of the lens element 32, including the lens 40, the flat surface 49 of the lens element 34, including the lens 46 have anti- reflective coatings (only one of which, i. e. , the coating layer 80 on the flat surface 49 is shown in Fig. 23). This coating 80 is index-matched with a glue layer 78, e. g. , a NOA-61 optical epoxy layer shown in Fig. 3, which may have a maximum thickness of about 4-5 mm. This improves optical coupling of the lens to the fiber and eliminates mechanical mismatch that may be caused by thermal deformations.

The end of the fiber 76 opposite to the lens assembly 30 is inserted into a ferrule 84 of another optical coupler 86 (Fig. 21), which connects the fiber 76 with an output optical fiber 88. As can be seen from this drawing, the ferrule 84 has a through opening 90. The end of the fiber 76 is inserted into one end of this opening, while an aspheric circular microlens 92 of a plate-like microlens is inserted with a tight fit into the opposite end of the opening 90.

The microlens element 94 is glued to the mating end face 96 of the ferrule 84 with a layer 98 of a UV-curable glue. The aforementioned end of the fiber 76 is located a certain distance from the aforementioned aspheric circular microlens 92. This distance ensures formation of a collimated light beam OB2 in a tubular separator 110 which is mentioned below.

When the fiber 76 is fixed in the ferrule 84 by a layer 102 of a UV-curable glue, the fiber end face should be in an exact location with respect to the microlens 92. If necessary, the exact positioning and fixation of the fiber end face can be facilitated by using an additional ferrule 103 and a layer 102 of the glue.

The flat end face of the microlens element 94 is coated with a mirror coating M1 which passes only a fraction, e. g. , about 10% of a selected narrow wavelength band of light incident on this mirror coating and reflects the remaining 90% of the selected wavelengths band of light through the fiber 76 back toward the laser. The 10%/90% ratio may vary to suit specific application and a desired power spectrum relation. For example, for red light the selected band may be of 635 nm 0.4 nm. The semiconductor laser diode 10 may be, e. g. , the one that generates light in the spectrum band of 635 nm 12. 5 nm (semiconductor laser diodes produced by Hitachi, Sony, Toshiba, Philips, etc. ). The mirror M1 will reflect approximately 100% of light except for the portion that corresponds to the wavelength of 635 nm 0.4 nm.

The flat rear end face of the microlens element 94 is glued via a layer 106 of a UV-curable glue to the front end face of the aforementioned tubular separator 110 having a central opening 112 of a diameter larger than the diameter of the fiber 76.

The flat front end face of another plate-like microlens element 116 is glued via a layer 118 of a UV-curable glue to the rear end face of a ferrule 122. A circular aspheric microlens 124, which is formed on the flat front side of the microlens element 116 inserted into a through opening 126 of the ferrule 122.

An output optical fiber 88 of the entire system is inserted into the end of the opening 126, which is opposite to the fiber 76. The end face of the fiber 88 should be located at a predetermined distance from the lens 124. In a real construction, positioning and fixation of end faces of respective fibers 76 and 88 are carried out so as to obtaining the maximum output light signal in the fiber 88.

Located on the left side of the laser diode 10 in Fig. 21 is an optical component of the system, which contains an extension unit 128 of the laser cavity L that is described later. By definition from Photonics Dictionary published in 1993 by the Publisher of Photonic Spectra Magazine, a laser cavity is an optical resonant structure, in which lasing activity begins when multiple reflections accumulate electromagnetic field intensity. It is difficult, however, to define a laser cavity in an optical system, which has many reflecting surfaces, which limit the area with lasing activity. Therefore, in the present patent application we define the laser cavity as a space from the mirror M1 to the Bragg grating 130. More specifically, the laser cavity extension unit 128 consists of a second anamorphotic objective 31 formed by a pair of microlens elements 33 and 35 with a spacer 37 between them. The second anamorphotic objective has the same construction and arrangement of parts as the first anamorphotic objective 30 described above. In other words, its end faces are flat and parallel to each other and are treated to a high degree of flatness in order to provide self-alignment and accurate coaxiality with the optical axis of the system during assembling. Similar to the fiber support and positioning system of the previously described couplings, the unit 128 has a cylindrical ferrule 67, which is centered in the groove 20 of the housing H coaxially with the rest of the optical system components.

A locking optical fiber 77 is inserted into a central through opening 71 of the ferrule 67 and is butt-connected to the flat rear end face 51 of the microlens element 33 with the butt connection described with reference to Fig. 23. A cylindrical aspherical microlens 41 of the microlens element 33 is snuggly fitted into the central opening 39 of the spacer 37. Similarly, a cylindrical spherical microlens 45, which has its longitudinal axis perpendicular to that of the microlens 41, is snuggly fitted into the opening 39 of the spacer 37 from the side opposite to the microlens element 33.

In fact, the assembly consisting of the microlens elements 33,35, spacer 37, and the ferrule 67 is identical to the assembly of microlens elements 32,34, etc. , which is mirror-image construction of the assembly locate on the left side from the laser diode 10. The left-side assembly has the same antireflective coatings and UV-curable glue layers as the right-side assembly, and therefore their description is omitted.

As shown in Fig. 21, the locking fiber 77 has a Bragg grating 130 written into the core of the optical fiber. The position of the Bragg grating 130 depends on a specific design and can be anywhere along the length of the locking fiber 77.

The applicant has successfully tested laser-diode assemblies of the invention with the different lengths of the fiber 77 within the range of 2 cm to 20 cm.

The free end of the locking fiber, behind the Bragg grating 130, is supported by a ferrule 131. It is understood that all ferrules 131,67, 68,84, and 122, as well as the spacers 31,36, and 110 have the same diameters, which are equal to the width of the groove 20 of the housing H. This ensures self- centering and alignment of the fiber-supporting elements and, hence, of the fibers themselves in the optical system of the invention.

The Bragg grating 130, the portion of the locking fiber 77 from the Bragg grating to the butt connection with the anamorphotic objective 31, the objective 31 itself, the space between the anamorphotic objective 31 and the rear end of the laser diode 10, the laser diode 10 itself, the space between the laser diode 10 and the anamorphotic objective 30, the anamorphotic objective 30 itself, the entire optical fiber 76, and the distance from the front end of the fiber 76 to the mirror coating M1 on the flat end face of the lens element 94 form an extended laser cavity L.

It can be seen that in contrast to the laser cavity L of the system disclosed in the earlier US Patent Application No.., light source, i. e. , the laser diode 10 is an intercavity element, which is located between the Bragg grating on one side and the laser mirror M1 on the other side. This means that the length of the laser cavity can be extended to a much greater degree than in the previously described construction. Another advantage is that there is no need to form a full-reflection mirror on the back facet of laser diode 10, or another source such as a semiconductor amplifier, or a superluminescent emitting diode. This is an important advantage since the laser diode works in an intensive temperature and light-power density mode which require that for reliability of operation the mirror coating on the laser be produced with an extremely high quality.

In the system of the present invention, the function of the aforementioned full- reflection mirror of the system, described in the aforementioned US Patent Application, is fulfilled by the Bragg grating 130. Along with the function of the full-reflection mirror, the Bragg grating 130 selects the linewidth and ensures optical power stability.

Bragg gratings are also known as distributed Bragg reflectors, which are optical fibers or other media that have been modified by modulating the longitudinal index of refraction of the fiber core, cladding or both to form a pattern. A fiber equipped with Bragg grating functions to modify the optical passband of the fiber (transmission characteristic) in such a way as to only transmits a narrow and controlled wavelength band. The distributed Bragg reflectors typically are"lossless"devices. In principle, the Bragg gratings can be used as light reflectors or as spectrum shape or mode converters.

A typical distributed Bragg reflector comprises a length of optical fiber including a plurality of perturbations in the index of refraction substantially equally spaced along the fiber length. These perturbations selectively reflect light of wavelength I equal to twice the spacing L between successive perturbations times the effective refractive index, i. e., t=2n L, where) is the vacuum wavelength and neff is the effective refractive index of the fiber for the mode being propagated. The remaining wavelengths pass essentially unimpeded. In the system of my invention, such a distributed Bragg grating 130 is used as a spectrum shape and mode converter for narrowing the spectrum bandwidth of the light radiated from the laser diode 10, as well as for stabilization of the output laser diode characteristics and for gaining the light energy which is resonated within the laser cavity L.

By selecting an appropriate periodic spacing L between successive perturbations in the fiber 77 with the distributed Bragg grating reflector 130, it becomes possible to select a mode, which is the most efficient for the operation of the semiconductor laser diode 10. In the system of the invention, such a mode is the one with the maximum intensity in the laser radiation spectrum. At the same time, the gain of the maximum intensity mode is accompanied by the suppression of the side modes of the spectrum.

Although only one antireflective coating 80 is shown on the end face 49 (Fig.

23), anti-reflective coatings (not shown) can be applied onto the end faces of optical fiber 77, of the microlens elements 33,35, 32,34, etc.

The optical system shown in Figs. 21-23 operates as follows : After the semiconductor laser 10 (Fig. 21) is activated, a diverged light beam, e. g. , of 635 nm 12.5 nm wavelength emitted by the laser 10, propagates in both directions, i. e. , toward the Bragg grating 130 and toward the output optical fiber 88.

The photons which propagate toward the Bragg grating 130 (Fig. 21) propagate through the anamorphotic objective 31 and are turned into a focused beam, which is coupled into the optical fiber 77. On its way, the light enters the distributed Bragg grating 130, which reflects the wavelength in the selected narrow bandwidth. A portion of the selected mode spectrum is reflected back to the laser diode 10. Immediately after initiation of the light- generation operation (fractions of nanoseconds), the system is self-adjusted to a mode operation in which the spectrum of the generated light will be readjusted to the narrow linewidth mode which will be further maintained due to operation of the Bragg grating 130.

In other words, the photons reflected from the Bragg grating 130 will propagate toward the mirror M1. After initiation of the laser diode 10, the process takes few cycles of photon reflections back and forth between the Bragg grating 130 and the mirror M1 (the cavity length), whereby the laser cavity enables light amplifications, i. e. , gain for the selected wavelength.

The intensified light of the selected mode then enters the microlens 92 of the microlens element 94 and passes to the lens element 116 via the mirror M1 and through the opening 112 of the spacer 110 to the microlens element 116.

The mirror M1 passes only a portion, e. g. , 75-99%, of the light in the selected wavelength band, e. g. , of 635 0. 4 nm, to the output fiber 88. The remaining portion of the light, e. g. 1 to 25%, is reflected back to the Bragg grating 130 via the aforementioned optical elements of the laser cavity L. In the case of the system of the invention, the optimum conditions were achieved at back reflection of 15-25%. When this reflected light enters the Bragg grating 130, the process of spectrum transformation and intensification of light of a selected wavelength with suppression of side modes is repeated. Thus, if 1 % of the light is reflected back for use in maximization of the gain of the laser system, this relatively weak feedback beam will not interfere with the main beam of the interest. In one of the designs tested by the applicant, the maximum gain was obtained when the emitter of the laser diode 10 was coated with a coating having reflectivity below 1%.

Similarly, the photons which are emitted from the laser diode in the direction opposite to the Bragg grating 130, in the direction of the output optical fiber 88, first pass through the lens element 32, the microlens 40 of which focuses this beam on the end face of the optical fiber 76. The focused beam is then propagates through the optical fiber 76 towards the mirror M1, which effects this beam toward the Bragg grating 130. On its way to the Bragg grating 130 the light beam is processed in the order reversed to steps in the direction of the mirror M1. The rest of the processing of this portion of the light beam which has initially been propagated towards the mirror M 1 is the same as has been described with regard to the light beam initially directed from the laser diode 10 to the Bragg grating.

Such an arrangement makes it possible to maintain high level of light radiation power on the selected frequency, which in the illustrated embodiment is the frequency of 6350. 4 nm. In combination with temperature control via a heat sink 21, it becomes possible to ensure long-term stability of the output light power with deviations not exceeding, e. g., 1%, or even lower than 0. 1%. Furthermore, the laser cavity with the external Bragg grating 103 may have an extremely long dimension, as compared to the length of a semiconductor diode chip. This allows not only obtaining of an extremely narrow linewidth, but also high stability of the frequency which is typical of laser systems with large external resonators.

The system of the invention has a simplified construction, assembling and adjustment by utilizing a three-functional component, which is the Bragg grating 130. These functions are frequency stabilization, narrowing of the line width, and partial light reflection for maximizing the gain of the system.

The embodiment of the invention shown in Figs. 21-23 is advantageous in that it allows the use not only a custom-designed laser but also a commercially produced light source. However, further simplification of the construction with elimination of one of optical couplings and one of optical fibers can be accomplished by means of an embodiment shown in Fig. 24. This embodiment, in general, is similar to that shown in Figs. 21-23 and differs from it by eliminating the output optical fiber 88 and the optical coupling (84, 86,122). In the embodiment of Fig. 4, the parts identical to those of Figs. 1-3 are designated by the same reference numerals with an addition of a prime.

An additional reference numeral 47 designates a rear end face of the microoptical lens element 32 Furthermore, the reflecting mirror M1 is transferred to the front facet 10a of the laser diode 10 The function of the output fiber 88 is transferred to a fiber 76 In compliance with the principle of the present invention, the laser diode 10 remains between the Bragg grating 130 of the cavity extension fiber 77 and the reflecting mirror M1 The length of the laser cavity L can be chosen without limitations, as in the previous embodiment. The principle of operation also remains the same and therefore is skipped from the description.

Fig. 25 illustrate an embodiment of the invention which is similar to one shown in Fig. 24 but differs from it only by the position of the reflecting mirror. Since both embodiments are very similar, in Fig. 25 the parts identical to those of Fig. 24 are designated by the same reference numerals but with two primes.

For example, the laser cavity L of the embodiment of Fig. 24 corresponds in Fig. 25 to the laser cavity L Furthermore, the description of the identical parts and of their operation is omitted.

The main distinction of the embodiment of Fig. 25 from the embodiment of Fig. 24 is that the reflecting mirror M1 is formed on the back end face 47 of the microoptical lens element 32 Thus, the laser cavity L is formed between the Bragg grating 130 and the reflecting mirror M1 The operation of the system of Fig. 25 is the same as of the system of Fig. 24.

Thus it has been shown that the invention provides a laser-diode device, which is characterized by a very narrows linewidth in a spectrum of light of a selected wavelength, an increased output signal/noise ratio, increased output light power at a selected narrow wavelength band, and stabilized frequency at the output. The device of the invention is suitable for use in microoptical systems with possibility of efficient assembling and alignment under mass production conditions. The invention also provides a method of stabilizing frequency and narrowing the linewidth of the spectrum of the light emitted from the laser device through external extended laser cavity. The invention simplifies the construction, assembling and adjustment by combining the functions of frequency stabilization, wavelength selection, and partial light reflection for maximizing the gain of the system in one optical component, which is a Bragg grating.

Although the invention has been described with reference to specific embodiments, it is understood that these embodiments were given only for illustrative purposes and that any changes and modifications with regard to shapes, designs, materials, and combinations thereof are possible, provided these changes and modifications do not depart from the scope of the patent claims. For example, the light source may comprise a superluminescent laser diode, a laser diode with an amplifier, etc. The housing H can be divided into two separate parts, one for the laser unit with the first coupling and anamorphotic lens assembly, and another one with the second coupling and the output fiber. This will allow individual temperature control optimal for separate units. The connection of the optical elements can be achieved by thermal fusion, rather than by adhesion. The mirror and the Bragg grating can be located in positions different from those described and shown in the illustrated embodiments, e. g. , the first mirror can be installed on a separate support behind the semiconductor laser diode and at a distance from this diode.

An opto-electronic interface module of the present invention is schematically shown in Figs. 26-29, where Fig. 26A is a sectional view illustrating coupling of the optical fiber with a miniature photodetector in accordance with the principle of the present invention, Fig. 27 is a simplified plan view of a unit consisting of the interface of the invention and a substrate with a hybrid circuitry of commercially produced electrical components, Fig. 28 is a sectional view similar to the one of Fig. 26A for an array-type interface, and Fig. 29 is a simplified block diagram illustrating electrical connections between the components of the device of the invention.

As shown in Fig. 26A, the opto-electronic interface module of the present invention, which hereafter will be referred to simply as a"device", consists of a microlens element 20 which has a convex microlens 22 and which is sandwiched between a glass tubularferrule 24 and a sensor-holding substrate 26 with a photodetector 28 such as a photodiode. The back side of the microlens element is designated by reference numeral 39. The sensor holding substrate 26 may comprise, e. g. , a silicon wafer type substrate with electric circuitry (for temperature sensing, impedance matching interface, thermoelectric cooling elements) applied, e. g. , by metallization, as well as with temperature sensors formed by photolithography for direct monitoring of temperature under low-temperature conditions, etc. The backside of the substrate 26 may be used as a ground shield for RF shielding. The entire optoelectronic assembly, consisting of the ferrule 24, the microlens element 20, and the photodetector 28 with the circuitry, etc. can be integrated into a package with a PC board with necessary electronics, e. g. , for low speed metro application.

Arrangement of the aforementioned components of a package are shown in Fig. 26B, which is a sectional view along the line 26B-26B of Fig. 26A. This is a simplified view, which is shown only as an example since many other arrangements are possible. In this drawing, the small circular area 29 inside the photodetector 28 designate an active planar zone of the photodetector made, e. g., of InP. Reference numerals 33a and 33b designate matching electroconductive stripes that connect the active zone 29 of the photodetector 28 with a power supply source and a pre-amplifier 33f, respectively.

Connection to the power supply is carried out via a wire hole 33c, while connection to the pre-amplifier is carried out via a capacitor 33d and a wire hole 33e. The linear stripes are shown conventionally since for impedance matching they may have any configuration such as serpentine, S-shaped, or other form. Symbol"G"designates a ground bus.

As seen in Fig 26B the pre-amplifier 33f can also be formed on the back surface of the photodetector 28'or on the substrate 26'.

As shown in Fig. 26A, an optical fiber 30 is inserted into a central opening 32 of the ferrule 24. The end face 34 of the optical fiber 30 is spaced from the nearest point of the microlens 22 at a distance"d"that ensures focusing of an optical beam I B onto the center of the active area of the photodetector 28.

The microlens element 20 can be made of an optical material such as glass, quartz, or an optical plastic and may have a thickness that depends on the location of the focal plane of the microlens 22 for focusing a light beam IB emitted from the end face 34 of the fiber 30. The microlens 22 formed on the side of the microlens element 20 that faces the ferrule 24 may be an aspheric circular microlens, a cylindrical microlens, or a lens of any othertype, provided that it projects from the plane of the microlens element 20. The microlens 22' should have a base diameter"D"equal to the diameter of the central opening 32 of the glass ferrule. The central opening of the glass ferrule is fit on the part of the microlens 22 which projects above the upper surface 37 of the microlens element 20 so that the ferrule is self-aligned and centered on the lens coaxially with the optical axis X-X of an optical fiber 31 inserted into the central opening 32 of the ferrule.

As shown in Fig. 26A, the buffer layer 23a of the optical fiber is stripped off and the cladding layer 33 is inserted into the central opening 32 of the ferrule.

For protection of the fiberfrom bending and breaking in the area of connection thereof to the ferrule 24, a rubber sleeve 23b can be fitted onto the buffer layer 23a. The end face of the buffer layer 23a is glued to the upper end face of the ferrule 24 by a glue layer 23c.

The base of the ferrule opening can have a flared edge to allow for fit on the <BR> <BR> <BR> @@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@ @@@@@@@@@@@@@@@@@@ optical axial alignment with a minimum air-gap between the ferrule end surface 36 and flat surface 37 of the lens element 20. This is important to allow for good and strong bonding between the ferrule 24 and lens element 20.

The base diameter"D"of the microlens and hence the diameter of the central opening 32 of the glass ferrule can be slightly, e. g. , by 1 j. m, greater than the diameter of the fiber cladding equal to 125 jj. m, if a standard optical fiber is inserted into the opening 32. Depending on the wavelength of the transmission, a typical fiber core of a single-mode fiber has a diameter within the range of 3 lim to 9 p. m. tn the case of a polarization-maintaining single- mode fiber, the characteristic transfer dimensions of the core also fall into the same interval of 3 jj. m to 9 m Less than 1 micron tolerance on the diameters of the ferrule opening 32 (which is typically of 126 pLm) and on the outer diameters of the fiber cladding and the base diameter"D"of the microlens should ensure tight fit of the lens and of the fiber inside the central opening 32. It is important for the end face of the ferrule 24 to have a high degree of flatness to ensure perpendicularity of the optical axis to the end face of the ferrule.

The optical fiber 30 can be fixed to the ferrule 24 by glue, e. g. , UV curable glue, or by means of YAG-laser welding.

Since the ferrule 24 is fit with its opening 32 onto the microlens 22, the latter functions as a centering and aligning element for the ferrule 24, so that after fitting onto the microlens with the end face 36 of the ferrule in contact with the surface of the microlens element 20, the longitudinal axis of the ferrule, and hence of the optical fiber 30, is oriented strictly perpendicular to the plane of the microlens element and hence coaxially with the optical axis X-X of the microlens 22.

The ferrule 24 is fixed to the microlens element by means of a layer 38 of glue, preferably, UV-curable glue, such as Norland 61 or equivalent available from the manufacturers.

For efficient coupling, the lower surface of the photodetector 28 is attached to the flat surface of the holder 28 via a thin layer 40a of a glue (preferably thinner than 5 µm). The microlens assembly (which includes the microlens 20, the ferrule 24, etc. ) is then attached to the upper flat surface of the holder 26 via a thin layer 40b (preferably thinner than 5 piu), which is optically matched to the lens element 20. Parallelism of the microlens element 20, substrate 26, and photodetector 28 to each other is ensured by utilizing spacers 27 and 29. These spacers have a calibrated height of about 160 um.

The thickness of the photodetector is about 150 yIm.

If necessary, the assembling can be carried out without the use of the spacers, since the surfaces of the holder 28 are produced with high parallelism, and the supporting surface of the photodetector 28 is relatively large (about 1 mm x 0.7 mm).

The photodetector 28 may be a photodiode. It may have an active area as small as 3 to 12 lim. It should be noted that the beam spot focused on the surface of the active area of the photodetector 28 has a diameter equal @@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@ @@@@@@@@@@@@@@ of the microlens 22 is located in the center of the aforementioned active area of the photodetector 28.

The assembling, focusing, and fixation of the aforementioned components of the optical unit shown in Figs. 26A and 26B will be now described with reference to Fig. 27A and 27B, which are simplified plan views of units made in accordance with two different embodiments of a device of the invention consisting of the substrate 26 (26') and a substrate 46 (46') with a hybrid circuitry which interconnects commercially produced electrical components such as a transimpedance amplifier 60 (60'), a digitization/clock generator unit 70 (70'), and an output digital amplifier 72 (72'). The difference between the embodiments of Figs. 2A and 2B consists in that in the case of Fig. 2A the trans-impedance amplifier 60 is formed on a substrate 46 with a hybrid circuitry, while in the case of Fig. 2B the trans-impedance amplifier 60'is formed on the substrate 26'in combination with photodetector 28'.

Since the assembling procedure for the arrangements of Figs. 27A and 27B are almost identical, the assembling will be further described only for the embodiment of Fig. 27A.

An electric pattern for electrical connections of the photodiode 28 to the trans- impedance amplifier 60 is formed by photolithography on the surface of the substrate 26. At the same time, the impedance matching stripes, such as 33a and 33b, are formed on the surface of the photodetector 28. The electrical components of the unit are connected to appropriate devices located on the back side of the substrate 26 via wire holes, such as 33e and 33c (Fig. 27B).

Then the photodetector28, such as a photodiode, is placed onto the substrate 26 to a marked position in which the output terminals 50 and 51 of the photodetector 28 are aligned to the terminals 52 and 53 of the trans- impedance amplifier 60 on the substrate 46 (Fig. 27A).

For high-frequency operation of the system, e. g. , with the frequency of about 40 GHz, the impedance on the output terminals 50 and 51 of the photodetector 28 must be impedance-matched to the input terminals 52 and 53 of the trans-irtipedance amplifier 60 and to input terminals 55 and 57 of the digitization unit 70 via the transimpedance amplifier 60. The high- frequency operation is also ensured due to the use of microstrips 50, 51,52, 53,55, and 57 between the components shown in Fig. 27.

Alignment of microstrips with the respective terminals and subsequent connections between the terminals, e. g. , in points 54 and 56, e. g., by YAG- laser welding or soldering, are carried out under a microscope or a computer- controlled vision system (not shown).

After connecting the photodetector-holding substrate 26 to the hybrid circuitry substrate 46, the electronics is subjected to DC and RF testing of performance characteristics of the interface in a special test chamber (not shown), and the electrical pulses converted from optical pulses by the photodetector 28 are modulated at the operating frequency. Once the stripline interconnections passed the test, the lens element 20 with the microlens 22 is positioned under a microscope observation onto the photodetector-holding substrate 26. In this case, a layerof UV-curable epoxy glue 42 is applied onto the surface of the cured glue layer 40 on the surface of the photodetector-holding substrate 26. This is a very thin layer of glue, and the optical coating on the lens flat surface is index matched to the glue to optimize optical coupling efficiency.

The alignment procedure consists in the following. The projection of the microlens 22 must be aligned with the position of the working area of the photodiode 28. In other words, the center of the microlens 22 must coincide with the center of the working area of the photodiode 28. Once the alignment is achieved, the components are interconnected by curing the glue layer 42. Some of the epoxy also covers the electric circuitry and thus protects it from humidity, dust, etc.

After connection of the microlens element 20 to the photodetector-holding substrate 26 is completed, the unit is again tested for operation. Once it passed the test, the glass ferrule 24 is positioned on the lens 22.

As has been describe above, the ferrule 24 fits with its flared of straight opening 32 onto the microlens 22 with a tight fit, so that the microlens 22 functions as a centering and aligning element for the ferrule 24. After fitting onto the microlens 22 with the end face 36 of the ferrule in contact with the surface of the microlens element 20, the longitudinal axis of the ferrule 24, and hence of the optical fiber 30. is oriented strictly perpendicular to the plane of the microlens element and hence coaxially @@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@ @@@@@@@@@@@@@@ 38 of a glue, e. g. , a UV-curable or heat-curable glue, is applied onto the outside perimeter of the ferrule in the area of contact of the ferrule 24 with the surface of the lens element 20, whereby the ferrule is glued to the lens element by UV radiation of the layer 38.

An optical fiber 30 is prepared for insertion into the ferrule 24 by stripping the fiber buffer (not shown), and cleaving the core 31 and cladding 33 flat. The treated end of the fiber 30 is then inserted into the central hole 32 of the ferrule 24.

The fiber 30 is inserted until the end face 34 of the optical fiber 30 touches the lens 22, and the fiber 30 is moved up by means of a micropositioning mechanism for a distance"d"required for focusing the beam IB emitted from the end face 34 of the fiber to the center F of the photodetector 28.

The above description related to an opto-electronic interface module consisting of a single optical fiber and a singie photodetector with an appropriate coupling and electrical connections. Fig. 28 shows an opto- electronic interface module that contains an array of photodetectors coupled to a plurality of optical fibers inserted into the centra ! openings of the ferrules also arranged into an array.

More specifically, the device of the embodiment of the invention shown in Fig.

28 has an array 80 of individual photodetectors 82a, 82b,.... 82n mounted on the surface 84 of a substrate 86. The substrate 86 supports a lens array 88 made of quartz, glass, etc. with individual microlenses 90a, 90b,... 90n formed on the surface 92 of the microlens array 88, e. g. , by photolithography. The pitch between the microlenses 90a, 90b,... 90n is equal to the pitch between the individual photodetectors 82a, 82b,.... 82n. The microlens array 88 is connected to the array 80 of individual photodetectors via a layer 94 of an indexed-matched material such as UV-curable glue. Reference numerals 96a, 96b,... 96n designate a plurality of glass or quartz ferrules self-aligned with the microlenses 90a, 90b,... 90n and containing optical fibers 98a, 98b,... 98n which may be connected to fibers, e. g. , of a multiple-fiber communication line (not shown).

The materials, functions of components, assembling, and alignment proceduresforindividual microlenses, photodetectors, and other components of the array-type interface shown in Fig. 28 are the same as have been described in connection with the embodiment of the invention shown in Figs.

26 and 27.

Fig. 29 is a simplified electric circuit of the system of Fig. 28. In Fig. 29, reference numerals 100a, 100b,... 100n designate transimpedance amplifiers connected between photodetectors 82a, 82b,.... 82n and a digital logic circuit 102. The transimpedance amplifiers 100a, 100b,... 100n are connected to output terminals of respective photodetectors 82b,.... 82n via stripline connectors 104a, 104a', 104b, 104b'... 104n, 104n'. Similarly, the transimpedance amplifiers 100a, 100b,... 100n are connected to the digital logic circuit 102 via RC circuits 106a, 106b,... 106n and stripline connectors 108a, 108b,.... 108n. Similar to the previous embodiment, all electrical components are mounted on respective substrates and their terminals are interconnected via electrical circuitry patterns formed by photolithography.

Fig. 30 illustrates another embodiment of the invention, where the optical and electrical components are arranged in a matrix form. For convenience of electrical connections, the matrices of photodetectors and optical components are formed by a plurality, e. g. , four arrays of the type described in the second embodiment. Since the optical matrix has the same contiguration as the matrix of the electrical components, only the latter is shown in Fig. 30. More specifically, a photodetector matrix 110 is formed by four arrays 112a, 112b, 112c, and112d of the type shown in Fig. 29 which for convenience of access are arranged on the peripheries of a square-shaped configuration with output terminals 114a, 114b, 114c, 114d, 114e,.... 114n of photodetectors 116a, 116b, 116c, 116d, 116e,.... 116n. Reference numerals 118a, 118b, 118c, and 188d designate arrays of transimpedance amplifiers. Each array 118a, 118b, 118c, and 188d is connected with a respective multiline digital logic circuit (only the digital logic circuit 120d is shown In Fig. 30). It is understood that the number of communication lines in each multilane logic circuit corresponds to the number of photodetectors in each photodetector array.

The interface module of the present invention can be produced in the form of a standard replaceable module of the type shown in Fig. 31 with pin/slot connections for interface with hybrid circuitry such as circuitry on the substrate 46 that consists of commercially produced electrical components. Fig. 31 is a three-dimensional view of the interface module 122 of the present invention.

The interface 122 consists of four photodetector arrays 124a, 24b, 124c, and 124d. Each photodetector array has the same construction as the one shown in Fig. 28. For example, the photodetector array 24c has an array of fertules 126a, 126b,.... 126n, fitted with a tight fit onto respective lenses (not shown), which in turn are connected with respective photodetectors (not shown).

Reference numerals 128a, 128b,... 128n designate output terminals of respective photodetectors. The entire module, including strip ! ine bridges, can be encapsulated into a molded plastic shell. This shell is shown in Fig. 30 by reference numeral 111. It encapsulates all optical and electrical components of the interface module, except for the optical fibers and the outputs of the photodetectors.

The principle of operation of the electro-optical interface of the invention is the same for all the embodiments described above. Therefore the operation of the device will be described only with reference to the embodiment of Figs. 26 and 27. A light signal is supplied to the optical fiber 30 from an optical data transmission system (not shown). A light beam IB is emitted from the end face 34 of the optical fiber 30 and propagates with divergence onto the surface of the microlens 22 of the microlens element 20. Since the thickness of the microlens element 20 is selected so that the beam is focused onto the surface of the back side 39 of the microlens element, the beam will also be focused onto the center F of the working area of the photodetector 28, which is in contact, via a very thin optically matched glue layer 42, with the surface 39. The photodetector 28 converts the optical signal into an electrical signal which is generated on the output stripline 50 and 51 of the photodetector 28. The electrical signal is sent through the sh iptine terminals 50 and 51 and the TIA 57 to the digital logic circuit (Fig. 20). The stripline terminals 50 and 51, as well as the stripline connectors 52, 53 and 55,57, etc. and the TIA ensure impedance matching between the interface module and the electric signal receiving bus (not shown).

Thus it has been shown that the present invention provides a simple, compact, and reliable opto-electronic interface which is suitable for mass production, can be produced in a miniaturized module form suitable for connection to a port of a personal computer, suitable for use in conjunction with high-speed voice data and video data transmission systems, facilitates focusing of optical beams emitted from the ends of optical fibers onto a very small photoreceiving areas, ensures automatic alignment of optical fibers with photodetectors during assembling, and functions as a combined mechanical holder of a fiber and a device for precision focusing onto the center of the photodetector.

Although the invention has been described and illustrated with reference to specific embodiments, it is understood that these embodiments should be construed as limiting the scope of application oF the invention and that any modifications and changes are possible, provided they do no). depart from the scope of patent claims. For example, the photodetector can be formed on a substrate together with the circuitry by means of planar technology. In the case of an array and matrix-type construction, flatness on the surface of the photodetector substrate mating with the surface of the lens substrate can be achieved by CMP planarization. The optical and electrical components may have different arrangements in arrays and matrices. The output terminals may have different configurations such as pins, holes, slits, etc. The interface module of the present invention can be used for interconnecting various optical data transmitting and electrical data receiving systems and can be utilized in personal computers, cellulartelephones. i~V sets, e tc. The stripline interconnection technique can be carried out by various methods, provided that they ensure matching of impedances on the input and output sides.